Stover Harvest – Safe and Sustainable With Good Resource Management Practices
Douglas L. Karlen1 and Stuart J. Birrell2
1USDA-Agricultural Research Service (ARS), National Laboratory for Agriculture & the Environment (NLAE)
2Department of Agriculture and Biosystems Engineering at Iowa State University (ISU)
Agriculture is being called upon to provide food, feed, fiber, and fuel now and for generations to come. To meet these critical needs our soil resources must not be degraded and our rural communities must be economically viable. One activity contributing to these efforts is a corn stover harvest study being conducted through a three-way partnership involving the USDA-Agricultural Research Service (ARS), Iowa State University (ISU), and POET LLC. Three years of field research have been completed evaluating seven corn stover management practices. Grain and stover yields have been measured and routine soil testing has been used to monitor for any changes associated with various stover management treatments. This report briefly summarizes what has been learned.
A field study was initiated on a 120 acre Clarion-Nicollet-Webster soil Association site near Emmetsburg, IA in 2008. Seven stover management treatments have been established. Each treatment is imposed on three replicated plots that are approximately five acres in size. They are: (1) conventional – no stover harvest; (2) cob only harvest; (3) harvest of plantmaterial other than grain (MOG) either directly or by direct-baling (Treatment 4); (5) a traditional two-pass harvest and baling operation, and (6 and 7) a high- (just below the ear) or low-cut harvest using a single-pass, dual stream biomass harvester developed based on a John Deere 9750 STS combine at ISU. Due to equipment differences, none of these treatments exactly duplicate what POET has asked their residue suppliers to do, but treatments 4 and 6 bracket the recommended practice of simply turning off the residue chopper/spreader and then baling the windrow that is created.
Two of the most critical questions being asked about stover harvest for bioenergy or any other use including animal feed or bedding are (1) will it reduce subsequent crop yields, and (2) will it degrade soil quality by increasing erosion, decreasing soil organic matter, depleting soil fertility, or having any other adverse environmental effects?
To quantify effects of stover harvest, the first step is to know your land. Stover should not be harvested from fields that have a high percentage of sloping areas or existing soil erosion problems. It is also important to have recent soil-test information to know the nutrient status of the field. You will also need to know some additional things about your corn hybrid such as the Harvest Index (HI). This can often be obtained from seed companies or determined by collecting on-site data to estimate the totalamount of crop residue that is being produced. For this study, we collected hand samples prior to machine harvest to estimate both the amount of above-ground biomass and the HI.Calculations based on both the measured amount of above-ground plant material and the HI determined from the hand samples and multiplied by the measured grain yields showed an average annual production of 3.7 tons/acre of stover. As long as corn grain is the predominant product being produced, the primary way to increase the potential amount of material available for harvest is to increase average corn yields. This point is clearly addressed in the up-coming revised “Billion Ton” report being prepared by people within the Department of Energy (DOE). Once the potential amount of stover available for harvest is known, dividing the actual amount collected by that value determines the harvest fraction.
Grain yields showed significant (P ≤ 0.1) seasonal effects (182, 162, and 155 bu/acre for 2008, 2009 and 2010, respectively), but this was not associated with the stover harvest treatments (Table 1) and the treatment by year interaction was not significant. Weattribute the 3-year average grain yield differences to a combination of highly variable soil fertility across the field and excessive rainfall in 2010.Soil samples were collected immediately after harvest, analyzed to monitor for changes due to stover harvest and used to assist with subsequent annual fertilizer recommendations. The three-year average values (Table 2) show no statistically significant (P ≤ 0.1) effects for soil pH, organic matter (SOM), or potassium (K). However, all four soil-test parameters showed highly significant replicate effects (data not presented). Based on soil pH, the small difference in average soil-test P is probably due to random placement of the high-cut STS plots rather than a stover harvest treatment effect. It is also important to understand that fertilizer recommendations are developed for various ranges of soil-test values so what this data shows is primarily a difference in philosophy regarding the amount of P and K fertilizer that was needed to address the variation in soils across the field. Interpreting the P and K data using one laboratory’s recommendations indicates all samples were in the optimum to high range while interpreting them according to another Laboratory’s guidelines ranks the values in the low to adequate range.One lesson learned from this study is that we should have used site-specific, differential fertilization or a higher overall average rate of fertilizer to overcome the spatial variability. Using ear leaf analysis at anthesis (pollination) shows that N was definitely limiting the crop in 2010 (2.97% versus 1.82% for 2009 and 2010, respectively). This was probablydue to the excessive amount of rainfall received that year. Low N was most likely a major reason for the lower grain yields in 2010, and was undoubtedly a much greater factor than any of the stover harvest treatments.
The relative grain yield for each treatment was determined by dividing the yield measurements for each treatment by the conventional yield (i.e. no stover harvest) for each replicate. Neither the three-year relative grain yield nor the average measured yield showed any statistically significant differences due to stover harvest treatment (Table 1).
Table 1. Average and relative corn grain yields as affected by various stover harvest treatments.
Stover Harvest Treatment / 2008 / 2009 / 2010 / 3-yearaverage yield / 3-year average relative yield
------bu/acre ------/ --- bu/acre --- / %
Conventional – no removal / 180 / 156 / 152 / 163 / 1.00
Cobs only / 178 / 153 / 151 / 161 / 0.99
MOG bulk collection / 186 / 153 / 152 / 164 / 1.01
Single-pass baling of MOG / 187 / 164 / 155 / 166 / 1.07
Two-pass baling / 177 / 167 / 157 / 167 / 1.03
STS High-cut / 179 / 171 / 140 / 163 / 1.01
STS Low-cut / 188 / 168 / 178 / 178 / 1.11
LSD(0.1) / NS / NS / NS
Table 2.Three-year average soil-test values as affected by various stover harvest treatments.
Stover Harvest Treatment / pH / SOM / P / K% / ---- ppm -----
Conventional – no removal / 6.86 / 4.43 / 24 / 164
Cobs only / 6.98 / 4.40 / 30 / 165
MOG bulk collection / 7.02 / 4.35 / 29 / 167
Single-pass baling of MOG / 7.13 / 4.52 / 29 / 177
Two-pass baling / 7.02 / 4.40 / 24 / 162
STS High-cut (just below ear) / 7.21 / 4.30 / 19 / 161
STS Low-cut (4 inch stubble) / 7.11 / 4.57 / 26 / 168
LSD(0.1) / NS / NS / 6 / NS
The quantity of stover harvested each year, the 3-year average, and the 3-year average percent of the potential above-ground, non-grain biomass that was collected are presented in Table 3. As expected, there were significant differences among the various harvest strategies. The 14% removal value for the cobs-only treatment is consistent with a multi-location corn cob studies. Halvorson and Johnson in the Agronomy Journalshowed that cobs accounted for 14 to 22% of the above-ground corn biomass in studies conducted at more than 20 locations. The 19% MOG and 22% single-pass baling of MOG represent two methods of collecting the cobs plus some of the upper leaf and stalk material. Both systems use a standard corn header but the machines are different. The bulk MOG treatment was performed using the STS machine with collection in a trailing wagon. The single-pass system had a baler attached directly to the combine so that material passing through the combine shakers would fall directly into the baler and be accumulated in large square bales. For the two-pass system, the first pass is with the combine which collects the grain and drops the stover behind the machine. For these plots, the residue was raked into a windrow and then baled. Differences in the aggressiveness of the raking operation is a major reason the amount of stover collected by baling was much greater in 2008 than in either 2009 or 2010 (Table 3). The STS High and Low cut treatments are single-pass, bulk collection treatments. Seasonal differences in the amount of material collected for these treatments reflects the degree of lodging associated with the crop as well as the ear height associated with the hybrid being grown. When there is substantial lodging, the “high and low” cut treatments are not very different because the corn head must be set first and foremost to capture all of the grain. These mechanical differences in addition to the aggressiveness of the raking operation explain why the “STS High Cut” system had a slightly higher average removal than the baling system. Overall, however, the estimated collection fraction for the two STS treatments are consistent with other field studies that we have conducted in Iowa and Minnesota.
Table 4 contains three pieces of information focused on N, P, and K. The first three columns show the average amount of these nutrients removed by the corn grain when averaged across all stover management treatments for the three years. The lower N removal for 2010 is consistent with the lower plant tissue N content discussed previously and attributed to excessive rainfall at this site in 2010. The P and K removal values reflect the amount of those nutrients taken off with the corn grain and are simply the product of the grain yield times the nutrient concentration.The middle three columns (Table 4) show the average additional amount of N-P-K that is removed when the stover is harvested. The important information from this data is that although there is an increase in nutrient removal when stover is harvested, the grain is still the predominant removal mechanism for N and P. For K, however, stover removal roughly doubles the amount of K removed and this will cause the soil test K values to fall if they are not monitored routinely. Finally, the last two columns in Table 4 show that soil-test values for P and K,when averaged across all treatments and years, did decline. We conclude that these soil-test changes confirm that higher rates of fertilizer should have been applied and that the decline in soil-test was caused more by soil variation across the field than by the stover harvest treatments.
Table 3.Seasonal and average corn stover yield and the average percent of above-ground, non-grain biomass collected as affected by various stover harvest treatments.
Stover Harvest Treatment / 2008 / 2009 / 2010 / 3-yearaverage yield / 3-year average percent collected
------tons/acre ------/ --- tons/acre --- / %
Conventional – no removal / 0.00 / 0.00 / 0.00 / 0.00 / 0
Cobs only / 0.46 / 0.66 / 0.50 / 0.54 / 14
MOG bulk collection / 0.67 / 0.90 / 0.59 / 0..72 / 19
Single-pass baling of MOG / --- / 0.87 / 0.80 / 0.83 / 22
Two-pass baling / 2.26 / 1.50 / 1.45 / 1.73 / 46
STS High-cut / 2.10 / 1.85 / 1.67 / 1.87 / 47
STS Low-cut / 2.24 / 2.50 / 2.45 / 2.40 / 60
LSD(0.1) / 0.08 / 0.10 / 4
Table 4.Seasonal grain and stover nutrient removal and subsequent soil-test P and K values.
Year / Grain removal (lb/acre) / Stover removal (lb/acre) / Soil-test (ppm)N / P / K / N / P / K / P / K
2008 / 95 / — / — / 14 / 3.1 / 24 / 35 / 197
2009 / 94 / 16 / 27 / 15 / 0.8 / 16 / 24 / 156
2010 / 85 / 24 / 31 / 15 / 2.5 / 15 / 19 / 148
LSD(0.1) / 5 / 2 / 2 / NS / 0.6 / 2 / 4 / 15
In summary, the three-year results of this collaborative project have shown that with good management, corn stover can safely and sustainably be harvested from fields similar to the one used for this research. Obviously, stover must not be harvested from areas subject to erosion and appropriate soil conservation practices should be applied to protect and improve those areas. Based on this study, we conclude that 1½ to 2 tons/acre of corn stover can safely be harvested from fields such as those represented by one used for this research. Routine soil testing and plant analysis should also be used to monitor effects of stover harvest and all farming operations. These are wise and economically beneficial practices and should be part of every farm management plan. For this specific location, it would be profitable to consider an increase in P and K fertilizer application rates. This recommendation may not be universal among all soil and crop management consultants depending upon the nutrient management philosophy they follow, but our experience on the Clarion-Nicollet-Webster soil association suggests that plant availability of both P and K is likely being affected by the high pH and Ca concentrations in the soil more so than by the slight increase in removal due to stover harvest.Finally, the results of this collaborative research project are also consistent with regional results associated with the USDA Renewable Energy Assessment Project (REAP) team and the Sun Grant Regional Partnership Corn Stover Teams for which we are integral partners.