Published in Agron J 100:690-695 (2008)
DOI: 10.2134/agronj2007.0204
© 2008 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
SOYBEAN
Year, Region, and Temperature Effects on the Quality of Minnesota's Soybean Crop
Seth L. Naeve* and
Sheri C. Huerd
Dep. of Agronomy and Plant Genetics, Univ. Minnesota, 411 Borlaug Hall, 1991 Upper Buford Circle, St. Paul, MN 55108. Research supported in part by the Minnesota Agricultural Experiment Station
* Corresponding author (naeve002{at}umn.edu).
 |
ABSTRACT
|
|---|
Bulk commodity soybean [Glycine max (L.) Merr.] can now be sourced with great specificity through rail and container purchases. The objective of this study was to determine whether analyses of farmer-grown soybean seed samples could detect significant regional differences in soybean quality traits. Through analysis of 2706 farmer volunteered soybean samples representing harvests from 2003–2005, we found significant year and region effects on seed quality. An independent variable, temperature during seed filling (Tsf) was created for each soybean sample by kriging local daily mean air temperatures from 135 weather stations. A region x year interaction was noted for protein and the sum of protein and oil concentrations, but using Tsf as a covariate eliminated this interaction. The variable Tsf correlated with average regional oil concentrations such that oil concentration increased at a rate of 6.6 g kg–1 °C–1. Overall, farmer produced and volunteered soybean samples provide an excellent means of identifying within-state soybean seed quality variation. Additionally, local climate data (Tsf) can be used to predict this variation in regional soybean seed quality.
Abbreviations: Tsf, temperature during seed filling
|
NOTES
|
|---|
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.
Received for publication June 9, 2007.
|
INTRODUCTION
|
|---|
INTEREST among international soybean purchasers, domestic soybean processors, and public and private soybean genetics suppliers remains high for detailed data regarding regional soybean quality variation. One reason may be that international purchasers continue to source soybean with origin specificity and stringent quality parameter requirements based on their needs. One tool making origin specificity increasingly important is containerized shipments of bulk soybean that can now be an economically viable transport mechanism (Ratajczyk, 2006). A second factor is the expanding biodiesel industry that is expected to use up to 10 million Mg yr–1 of soybean by 2015 (Urbanchuk, 2006). As of 31 Jan. 2007, 77 biodiesel production plants were under construction, while eight more were expanding (National Biodiesel Board, 2007). This expansion will lead to tight supplies of soybean oil (Carlson, 2006), and will increase the need for defining specific locations where crush and biodiesel production margins may be maximized. The increased disparity in value between protein and oil by the end user will cause soybean crushers to actively seek out regions producing high relative oil concentrations for plant placement.
A final factor increasing interest in regional soybean quality data is the anticipated introduction of new high-value soybean varieties that will require careful placement on the landscape to minimize risk and maximize profits for producers and others in the supply and utilization chain. Early examples are low-linolenic (Hammond and Fehr, 1983; Ross et al., 2000), and mid-oleic (Burton et al., 2006; Oliva et al., 2006) soybean varieties that produce oils that have a greater oxidative stability and do not require partial hydrogenation to prevent rancidity. Companies supplying new soybean varieties to the marketplace are very interested in geographic variation in seed quality so that they might best exploit specific oil or protein traits.
Regional quality variation of bulk commodity soybean has been characterized (Breene et al., 1988; Hurburgh et al., 1990; Hurburgh, 1994; Brumm and Hurburgh, 2006). Mounts et al. (1990) and Hurburgh et al. (1990) quantified national differences in the quality of export soybean from the United States, Brazil, and Paraguay and Argentina. Hurburgh et al. (1990) provide the most comprehensive view of regional protein and oil variation within the United States. Their work demonstrated that western and northern Corn Belt states produce commodity soybean with lower protein concentration. However, significant year to year variability occurs with protein (Hurburgh et al., 1990) and oil (Brumm and Hurburgh, 2006). Domestic and international purchasers have been aware of these natural regional differences, and have made purchasing decisions based on these patterns. Selective purchases of soybean by region would potentially affect regional soybean stocks. Differences in regional soybean stocks help to explain the fluctuations of local price vs. Chicago Board of Trade Price (i.e., basis).
Many factors have been used to explain regional protein and oil variation including latitude (Breene et al., 1988; Dardanelli et al., 2006), precipitation (Maestri et al., 1998), temperature (Howell and Cartter, 1958; Piper and Boote, 1999; Thomas et al., 2003; Yaklich and Vinyard, 2004), extreme temperature (Wolf et al., 1982; Gibson and Mullen, 1996), temperature and drought (Dornbos and Mullen, 1992), soil fertility (Haq and Mallarino, 2005), and tillage and rotations (Pedersen and Lauer, 2003; Temperly and Borges, 2006). While temperature effects on oil concentration may be the best characterized of any environmental factor/quality trait interaction, a definitive model describing this interaction has not been developed. Greenhouse and growth chamber studies have provided a wide range of results that depend greatly on the temperature range examined. Three groups (Gibson and Mullen, 1996; Pazdernik et al., 1996; Thomas et al., 2003) used average daily temperature ranges of about 22°C to as much as 39°C. All three groups found decreased oil concentrations with higher temperatures. Alternatively, Howell and Cartter (1958) found that oil concentration increased by about 4.5 g kg–1 °C–1 across a temperature range of 21 to 29°C. Wolf et al. (1982) found that oil increased by 3.3 g kg–1 °C–1 from 16 to 31°C. Published reports of in-field examinations of temperature effects on soybean quality are few. Yaklich and Vinyard (2004) determined that August and September temperatures best predicted oil concentration in soybean grown in three locations between 40 and 42° N lat. in the central United States. Kane et al. (1997) planted soybean varieties with relative maturity ratings of 00 to IV at four planting dates at one site in Kentucky (38° 7'N lat.). They found Tsf (R5–R7; Fehr and Caviness, 1977) to be correlated with seed oil concentrations in five of six varieties (r = 0.52–0.66). Piper and Boote (1999) provide a very thorough analysis of 20 check varieties grown in the Uniform Variety Trials at more than 60 locations from 1970 to 1990. They found that a quadratic model best explained temperature effects on soybean oil concentration (best-adjusted r2 = 0.24). The slope of this model ranged from 8.1 g kg–1 °C–1 at 15°C to 2.1 g kg–1 °C–1 at 25°C.
In light of the known variation in soybean oil and protein concentration across the Corn Belt, little information exists on variation within a state. The objectives of this study were (i) to detect significant regional soybean composition variation within a single state (Minnesota) through analysis of farmer volunteered grain samples, (ii) to quantify year to year variability within a region, and (iii) to determine whether air Tsf could explain any of these putative effects.
|
MATERIALS AND METHODS
|
|---|
Soybean sample kits were mailed to approximately 5000 Minnesota soybean producers each August of 2003, 2004, and 2005. Sample kits were distributed throughout the state by county in proportion to the soybean production area in the previous year (USDA-National Agricultural Statistics Service, 2002–2004), so that one kit was mailed for each 567 ha of soybean. Sample kits contained a letter explaining the project, a 125 x 200-mm plastic sample bag with a response questionnaire label, and a postage-paid return Tyvek envelope (DuPont, Wilmington, DE). Producers were asked to sample harvested grain from the variety that they planted on the largest number of hectares to ensure that sample results best represented the county, region, and state commodity soybean crop. Producers indicated the zip code of the town nearest the farm where the sample was taken, the soybean variety sampled, and answered several questions regarding field history, pest issues, and weed/pest controls used. Of the volunteered information, only location data were used in this paper.
Production regions were based on the National Agricultural Statistical Service Districts (USDA-National Agricultural Statistics Service, 2003–2005). Of nine regions, two (Northeast and North Central) were dropped due to a lack of soybean production. These regions are dominated by forests and most agricultural land is devoted to livestock and grazing. The North Central district produced soybean on fewer than 8300 ha during this 3-yr period, primarily from a single county, Hubbard. This county was therefore included in the Northwest region. Overall, 623, 952, and 1131 samples in 2003, 2004, 2005, respectively, were returned with an adequate quantity of soybean and written information for analysis. Soybean samples were processed for protein and oil using near infrared spectroscopy (NIRS) technology. A Foss full scanning 6500 monochrometer fitted with NIRS equations developed by the University of Minnesota using ISIscan software (Infrasoft Intl., State College, PA) and validated by Caltest (Clifton Park, NY) was used.
Each sample was assigned an approximate latitude and longitude value based on the zip code provided by the producer. These latitude and longitude values were, in turn, used to assign values to each sample for county, region, and weather data across Minnesota (between 43° and 49° N lat.; 89° and 96° W long.). Minnesota has 127 weather stations; eight stations near borders of neighboring states provided additional data. Stations are evenly spread statewide. Corresponding temperature data from the National Weather Service was compiled by the University of Minnesota Climatology Office using mean daily air temperatures between 15 August and 14 September for each year for each weather station. These dates were chosen as the normal soybean seed-filling period (R5-R7) across Minnesota. Prediction maps of mean daily temperatures were kriged in ArcGIS using Ordinary Kriging (ESRI, Software Version 9; ArcView) for the entire state. Kriging is an interpolation method that allows for examination of spatial autocorrelation of data. Thus, a local temperature during seed-filling variable (Tsf) was estimated for each soybean sample submitted (see below).
The seed-filling period used in this study (15 August–14 September) is based primarily on personal observation of the average R5 to R7 period in Minnesota across years. No published work was found to establish the variation of seed-filling period across regions with adapted varieties produced at each site. The National Agricultural Statistics Service (NASS) does provide data on percent of field at "pod set" and "leaf drop" by state (USDA-National Agricultural Statistics Service, 2007). NASS data indicate that pod set is initiated across states over a wide range in time (typically 30 d or more for 10–95% pod set), but initiation periods are similar between neighboring states. For example, 50% pod set in Iowa and Minnesota differ by <5 d, based on 2002–2006 data (USDA-National Agricultural Statistics Service, 2007). Leaf drop occurs simultaneously in these two states. Seed filling in this study is a best estimate for these samples that differ in variety, maturity, and planting date.
Soybean sample zip codes were used as a map overlay on the Tsf prediction map, such that each soybean sample location was assigned a Tsf in each year. The estimated Tsf and corresponding protein and oil concentration data were used for statistical analysis in SAS v. 9.1.3 (SAS Institute, Cary, NC). Year, region, county, and Tsf were treated as independent variables for all analyses. All ANOVAs were performed using PROC GLM. Variables were compared with LSD when significant. Sources of variation and means were declared significant at P < 0.05. Data from this study were also analyzed as a weighted dataset, whereby seed quality and Tsf were weighted by actual planted area of counties rather than by number of samples returned. Region values were determined by zonal statistics in ArcView, and analyzed in SAS as above. Because within-year response rate by county closely matched production area within those counties, differences between weighted and nonweighted analyses were marginal. Only the nonweighted data are presented.
|
RESULTS
|
|---|
The 3-yr survey period provided a wide range of climate variation by which to examine year and region effects on seed quality (Table 1
). The final year of the study, 2005, was a near normal year throughout August and September, with average temperatures and some minor, localized drought conditions occurring across the state. By comparison, 2003 tended to be dry and warm throughout the seed-filling period across Minnesota, and the month of August was the fourth driest on record with above average temperatures. September temperatures and rainfalls were more moderate. The second year, 2004, was much the reverse of 2003, with August 2004 being the coldest August on record and having above average rainfall. As with September 2003, the overall climate in September of 2004 was more moderate.
View this table:
[in this window]
[in a new window]
|
Table 1. Year, region, and region within year means for temperature at seed filling (Tsf), yield, protein, oil, protein + oil, and protein to oil ratio for all farmer samples returned for a 2003–2005 survey of soybean in seven Minnesota regions.
|
|
Table 1 provides estimates of total area devoted to soybean production and average yields by year, region, and region within year (USDA-National Agricultural Statistics Service, 2003–2005). It displays the number of samples returned by year and region as well as Tsf, average protein, oil, the sum of protein and oil concentrations (P+O) and the ratio of protein to oil (P to O) for those returned samples. The P+O serves as an index for processed value of commodity soybean. The P to O ratio provides an index for evaluating an environment's relative ability to produce either protein or oil. Sample numbers correlated well with county and regional soybean production within years (data not shown), indicating that regional and statewide average values for protein and oil concentrations reasonably reflect the quality of the crop as a whole at the region and statewide level.
There were significant differences among years for protein, oil, P+O and P to O (Table 1). Low protein levels were noted in 2005, while 2003 produced relatively high oil concentrations in Minnesota. The low protein concentrations observed in 2005 led to low P+O values, while 2003 produced soybean with the lowest P to O ratio of the study period due to the high oil values that year.
Inter-region differences in quality traits across years were not as large as analogous statewide values across years (Table 1). Despite an enormous range in background soybean genetics, whereby more than 260 variety names were volunteered yearly, sample numbers were sufficiently large to allow detection of statistically significant differences among regions for all traits. Southeast and East Central Minnesota produced crops with the highest protein concentration across this period, however, due to low total soybean production and correspondingly low sample numbers (41 of 2707 samples over 3 yr), East Central results should be interpreted with caution. The lowest protein soybean crops were produced in the opposite corner of the state (Central, West Central, and Northwest). Numerical differences in regional oil concentrations tended to be small except for the Northwest region where oil was very depressed relative to the other regions. Southern regions produced crops with higher P+O values than all regions except East Central. Protein to oil ratios among regions appeared to be driven primarily by oil concentration with Northwest Minnesota producing the highest P to O ratio and South Central and West Central the lowest. When analyzed by year and region (Table 2
), a significant interaction was found for protein and P+O. Oil concentration and P to O were affected by year and region without a corresponding interaction.
View this table:
[in this window]
[in a new window]
|
Table 2. Top section: ANOVA for soybean quality traits from seven regions across Minnesota in 2003–2005. Bottom section: same ANOVA as top section using temperature during seed filling (Tsf) as a covariate.
|
|
Incorporating Tsf into the statistical model affected year and year x region effects. When year, region, and year x region effects were adjusted for Tsf and Tsf interactions, the year x region interactions were no longer significant for protein and P+O. Likewise, Tsf explained enough of the quality variation noted that year effects were no longer significant for any traits. However, a significant Tsf x region effect was noted for all traits measured. Across the 3-yr study period, regions produced soybean crops with a range of values for quality traits, however, Tsf did affect regions consistently.
Coefficients of determination for Tsf effects on the four quality traits measured are shown in Table 3
. Across years, all regions showed significant correlations between Tsf and oil concentrations, with r values from 0.49 to 0.66 (r2 = 0.24–0.43). These consistent correlations were clearly dependent on the large year-to-year variation in temperature experienced by all regions. Correlations of Tsf and oil concentrations within years were much smaller, with r values of 0.15 to 0.36 (r2 = 0.02–0.13). Across years, Tsf was also significantly correlated with P+O and P to O, but values were smaller compared to the Tsf–oil relationship. With little overall Tsf effect on protein noted, the positive effect of Tsf on oil was the primary driver behind the positive and negative correlations between temperature and P+O and P to O ratio, respectively.
View this table:
[in this window]
[in a new window]
|
Table 3. Slope, intercept and r2 values from the regression of temperature during seed filling (Tsf) on soybean trait data from seven regions across Minnesota in 2003–2005.
|
|
Table 3 also provides linear regression models for all significant Tsf and quality trait relationships. It is clear that the Tsf x region interaction for oil concentration identified in Table 2 is primarily driven by the East Central region where the linear regression identifies Tsf on oil to be 4.2 g kg–1 °C–1. By comparison, all other regions showed slopes of 6.0 to 7.3 g kg–1 °C–1. Likewise, the Tsf x region interaction for protein concentration noted in Table 2 is likely due to the Northwest and Southwest regions where Tsf was negatively correlated with protein, while other regions were not correlated. The sum of protein and oil tended to be less responsive in East Central Minnesota where the oil response was low, and in the Northwest where a strong negative correlation between Tsf and protein was noted.
|
DISCUSSION
|
|---|
Regional variation for protein concentration was confounded by year effects within the state of Minnesota, while oil concentration varied greatly without a year x region interaction. While year to year variation in average oil levels is large, regions within Minnesota tend to produce soybean crops with oil concentrations of similar rank order across years. Reasonable estimates of soybean oil concentrations from a region should allow an estimate of soybean oil concentrations produced in other regions based on these rank order effects.
Temperature at seed filling (Tsf) effects explained much of the year to year and region to region variation found during this study period. Adding Tsf into the statistical model removed year effects from all terms examined. However, Tsf x region interactions were noted for all traits. A subtle regional sensitivity to Tsf accounts for this interaction. Regional differences in planted soybean varieties, length of the growing season, and susceptibility to drought and other environmental stresses are potential causes of this variation. Temperature at seed-filling (Tsf) effects on oil were significant across all major soybean producing regions of the state.
A fundamental question addressed by Yaklich and Vinyard (2004) and Piper and Boote (1999) is whether seed quality traits may be predicted based on late season ambient temperatures and past performance within a region. This research indicates that oil concentration may be reasonably well predicted based on late season temperatures within a region. Figure 1
shows the relationship between regional temperatures from 15 August through 14 September by year and average regional oil concentration as determined by this survey within those regions and years. Plotting average regional oil concentration vs. average regional temperatures provided a linear regression with slope of 6.6 g kg–1 °C–1 and an r2 value of 0.84 across a 15.5 to 21°C temperature range. While Piper and Boote (1999) found that the temperature/oil relationship of 20 varieties (of maturity groups 00–VII) grown from 29.4°N to 47.5°N lat. was best fit by a quadratic model, they found that linear regressions fit individual varieties well. Varieties adapted to the Upper Midwest showed a larger oil response to temperature compared to varieties adapted to Southern U.S. latitudes. Their seven varieties that match maturities of varieties commonly produced across the state of Minnesota (MG 00–II) provided an average slope of 5.7 g kg–1 °C–1. Southern adapted varieties (MG V–VII) had an average slope of 2.9 g kg–1 °C–1. Although the small number of very old varieties was grown across a very wide range of environments, the resulting temperature/oil relationship found by Piper and Boote (1999) for northern genotypes was very similar to the one we described here, where samples primarily represented contemporary, elite commercial varieties produced exclusively in their adapted zone.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1. Linear regression of the average regional ambient temperature with average oil concentration of samples from a 2003–2005 survey of the quality of soybean in seven Minnesota regions. Values are based on an average of locally estimated temperatures for each sample analyzed vs. the average oil concentration of corresponding samples.
|
|
While the regional relationship shown in Fig. 1 does not reflect a model predictive of soybean seed oil concentrations for individual Minnesota farmers based on local temperature, it does provide a reasonable prediction of oil concentrations of a bulk commodity soybean pool across the broader geography of Minnesota. Moreover, the observed relationship between temperature and quality of farmer produced samples substantiates results found in an elegant study of old varieties grown outside their adapted zone (Piper and Boote, 1999). Additionally, our results parallel the findings of some notable historical growth chamber and greenhouse work that was conducted across normal air temperatures for northern U.S. soybean production areas where oil concentration increased 4.5 g kg–1 °C–1 and 3.3 g kg–1 °C–1, respectively (Howell and Cartter, 1958; Wolf et al., 1982).
|
CONCLUSIONS
|
|---|
This study demonstrates the ability to use farmer-grown soybean samples to estimate the seed quality of the overall commodity pool on a statewide basis, and to determine regional differences in these quality traits. While no attempt was made to separate genotypic from environmental influences, a significant regional difference in soybean quality within the state of Minnesota was found. Purchasers who are interested in oil concentration alone should consider Southwest and South Central Minnesota sourced soybean to have equal or higher oil concentration than more northerly regions in most years. These southerly regions may be well suited for construction of crushing plants and biodiesel refineries. Regional oil concentrations were affected by Tsf during late August and early September at a rate of 6.6 g kg–1 °C–1.
|
ACKNOWLEDGMENTS
|
|---|
This project was made possible only through the financial generosity of the Minnesota Soybean Research & Promotion Council. We thank Tracy O'Neill for her expert technical assistance, and Jill Miller-Garvin for her critical review.
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.
|
REFERENCES
|
|---|
- Breene, W.M., S. Lin, L. Hardman, and J. Orf. 1988. Protein and oil content of soybean from different geographic regions. J. Am. Oil Chem. Soc. 65:1927–1931.[Web of Science]
- Brumm, T.J., and C.R. Hurburgh. 2006. Changes in long-term soybean compositional patterns. J. Am. Oil Chem. Soc. 83:981–983.[Web of Science]
- Burton, J.W., R.F. Wilson, G.J. Rebetzke, and V.R. Pantalone. 2006. Registration of N98–4445A mid-oleic soybean germplasm line. Crop Sci. 46:1010–1012.[Free Full Text]
- Carlson, R.W. 2006. Prospects for expansion of the soy-based biodiesel industry in Minnesota. Available at www.auri.org/research/diesel/pdfs/Executive%20Summary%20Bio-Diesel%20Study%20December%2006.pdf (accessed 6 June 2007; verified 11 Jan. 2008).Agric. Utilization Res. Inst., Crookston, MN.
- Dardanelli, J.L., M. Balzarini, M.J. Martinez, M. Cuniberti, S. Resnik, S.F. Ramunda, R. Herrero, and H. Baigorri. 2006. Soybean maturity groups, environments, and their interaction define mega-environments for seed composition in Argentina. Crop Sci. 46:1939–1947.[Abstract/Free Full Text]
- Dornbos, D.L., and R.E. Mullen. 1992. Soybean seed protein and oil contents and fatty acid composition adjustments by drought and temperature. J. Am. Oil Chem. Soc. 69:228–231.[Web of Science]
- Fehr, W.R., and C.E. Caviness. 1977. Stages of soybean development. Spec. Rep. 80. Iowa Agric. Home Econ. Exp. Stn., Ames.
- Gibson, L.R., and R.E. Mullen. 1996. Soybean seed composition under high day and night growth temperatures. J. Am. Oil Chem. Soc. 73:733–737.[CrossRef][Web of Science]
- Hammond, E.G., and W.R. Fehr. 1983. Registration of A5 germplasm line of soybean. Crop Sci. 23:192–197.[Free Full Text]
- Haq, M.U., and A.P. Mallarino. 2005. Response of soybean grain oil and protein concentrations to foliar and soil fertilization. Agron. J. 97:910–918.[Abstract/Free Full Text]
- Howell, R.W., and J.L. Cartter. 1958. Physiological factors affecting composition of soybeans: II. Response of oil and other constituents of soybeans to temperature under controlled conditions. Agron. J. 50:664–667.[Abstract/Free Full Text]
- Hurburgh, C.R. 1994. Long-term soybean composition patterns and their effect on processing. J. Am. Oil Chem. Soc. 71:1425–1427.[Web of Science]
- Hurburgh, C.R., T.J. Brumm, J.M. Guinn, and R.A. Hartwig. 1990. Protein and oil patterns in U.S. and world soybean markets. J. Am. Oil Chem. Soc. 67:966–973.[Web of Science]
- Kane, M.V., C.C. Steele, L.J. Grabau, C.T. MacKown, and D.F. Hildebrand. 1997. Early-maturing soybean cropping system: III. Protein and oil contents and oil composition. Agron. J. 89:464–469.[Abstract/Free Full Text]
- Maestri, D.M., D.O. Labuckas, J.M. Meriles, A.L. Lamarque, J.A. Zygadlo, and C.A. Guzman. 1998. Seed composition of soybean cultivars evaluated in different regions. J. Sci. Food Agric. 77:494–498.[CrossRef][Web of Science]
- Mounts, T.L., J.M. Snyder, R.T. Hinsch, A.J. Bongers, and A.R. Class. 1990. Quality of soybeans in export. J. Am. Oil Chem. Soc. 67:743–746.[Web of Science]
- National Biodiesel Board. 2007. Biodiesel production plants under construction or expansion (Sept. 7, 2007). Available at www.biodiesel.org/buyingbiodiesel/producers_marketers/ProducersMap-Construction.pdf (accessed 6 June 2007; verified 11 Jan. 2008).Natl. Biodiesel Board, Jefferson City, MO.
- Oliva, M.L., J.G. Shannon, D.A. Sleper, M.R. Ellersieck, A.J. Cardinal, R.L. Paris, and J.D. Lee. 2006. Stability of fatty acid profile in soybean genotypes with modified seed oil composition. Crop Sci. 46:2069–2075.[Abstract/Free Full Text]
- Pazdernik, D.L., S.J. Plehn, J.L. Halgerson, and J.H. Orf. 1996. Effect of temperature and genotype on the crude glycinin fraction (11S) of soybean and its analysis by near-infrared reflectance spectroscopy (Near-IRS). J. Agric. Food Chem. 44:2278–2281.[Web of Science]
- Pedersen, P., and J.G. Lauer. 2003. Soybean agronomic response to management systems in the Upper Midwest. Agron. J. 95:1146–1151.[Abstract/Free Full Text]
- Piper, E.L., and K.J. Boote. 1999. Temperature and cultivar effects on soybean seed oil and protein concentrations. J. Am. Oil Chem. Soc. 76:1233–1241.[CrossRef][Web of Science]
- Ratajczyk, C. 2006. Targeting competitive market channels for soy. Available at http://ussoyexports.org/news/key_issues/container_logistics.pdf (accessed 6 June 2007; verified 11 Jan. 2008). U.S. Soybean Export Council, St. Louis, MO.
- Ross, A.J., W.R. Fehr, G.A. Welke, and S.R. Cianzio. 2000. Agronomic and seed traits of 1%-linolenate soybean genotypes. Crop Sci. 40:383–386.[Abstract/Free Full Text]
- Temperly, R.J., and R. Borges. 2006. Tillage and crop rotation impact on soybean grain yield and composition. Agron. J. 98:999–1004.[Abstract/Free Full Text]
- Thomas, J.M.G., K.J. Boote, L.H. Allen, M. Gallo-Meagher, and J.M. Davis. 2003. Elevated temperature and carbon dioxide effects on soybean seed composition and transcript abundance. Crop Sci. 43:1548–1557.[Abstract/Free Full Text]
- Urbanchuk, J.M. 2006. Contribution of the biodiesel industry to the economy of the United States. Available at www.nbb.org/resources/reportsdatabase/reports/gen/20060930_gen373.pdf (accessed 6 June 2007; verified 11 Jan. 2008). Natl. Biodiesel Board, Jefferson City, MO.
- USDA-National Agricultural Statistics Service. 2002–2005. Minnesota annual statistical bulletin. Available at www.nass.usda.gov/Statistics_by_State/Minnesota/Publications/Annual_Statistical_Bulletin/index.asp (accessed 6 June 2007; verified 23 Jan. 2008). USDA-NASS, Washington, DC.
- USDA-National Agricultural Statistics Service. 2007. Crop production. Available at http://www.nass.usda.gov/QuickStats/index2.jsp (verified 23 Jan. 2008). USDA-NASS, Washington, DC.
- Wolf, R.B., J.F. Cavins, R. Kleiman, and L.T. Black. 1982. Effect of temperature on soybean seed constituents: Oil, protein, moisture, fatty acids, amino acids, and sugars. J. Am. Oil Chem. Soc. 59:230–232.[CrossRef][Web of Science]
- Yaklich, R., and B. Vinyard. 2004. A method to estimate soybean seed protein and oil concentration before harvest. J. Am. Oil Chem. Soc. 81:1021–1027.[Web of Science]
This article has been cited by other articles:
|
|
|
|
 
C. Carrera, M. J. Martinez, J. Dardanelli, and M. Balzarini
Water Deficit Effect on the Relationship between Temperature during the Seed Fill Period and Soybean Seed Oil and Protein Concentrations
Crop Sci.,
May 11, 2009;
49(3):
990 - 998.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
