The Michigan Corn Stover Project — Part 4: Stover harvest implications
On-farm research sheds light on what impacts corn stover harvesting will have on farming operations in Michigan.
The Michigan Corn Stover Project was a collaborative effort at Michigan State University Extension to investigate the uses of corn stover and potential impacts of harvest in Michigan. This effort was comprised of on-farm and small scale research conducted across Lower Michigan. It included a bale storage study, integration of a cover crop, a cattle feeding study and the impact of stover removal on yield of the subsequent crop.
The objectives of the corn stover harvest study were to determine whether harvesting stover would impact crop yield the following season, estimate the amount of stover removed compared with non-harvested plots and quantify the amount of nutrients (N, P2O5 and K2O) that would be removed per dry ton of stover.
Three locations (Photo 1) were chosen in 2014—two had been in a corn-corn rotation and one in a corn-soybean rotation. Sizes of the study fields ranged from 10–16 acres with plot widths of 24–72 rows. Corn was planted on 30-inch centers at corn-corn sites and 20-inch centers at the corn-soybean site. All fields were planted to corn in 2014, and stover was harvested from half of eight plots, randomly ordered, at each site. The study was replicated in 2015 and 2016; however, due to crop rotation and weather-related harvest problems, stover was harvested in only two years at two of the sites.
The impact of harvesting stover on grain yield the subsequent year is summarized in Table 1. Variability within field, among sites, and across years was high, and no statistical differences were detected between plots where stover had been harvested and where it had not been harvested. Crop yield was 7 percent higher in harvested plots compared with non-harvested plots when averaged across locations and years. Previous studies have shown some evidence that reducing the amount of corn stover can increase grain yields the following year, and although that was not found to be the case statistically in this three-year study, several plots did have numerically higher yields the following year.
Table 1. Crop yields (bushels per acre) in plots with and without stover harvest. | |||
---|---|---|---|
Site |
Harvested (h) |
Not harvested (nh) |
% change (h-nh)/h) |
A |
154 |
140 |
9 |
B |
113 |
91 |
18 |
C* |
139 |
145 |
-6 |
Average |
135 |
125 |
7 |
*Soybean planted in 2015 at site C.
The amount of residue cover remaining after baling was 13 percent less than in strips where no stover was harvested, which was statistically significant (P<0.01), when averaged over sites and years (Table 2). A Cornrower head was used to chop and windrow the stover at sites A and B. However, a separate stalk chopper/windrower was used at site C, which was not as efficient in collecting stover, so differences between plots could not be detected statistically (P=0.11). How much stover to leave behind to avoid a reduction in soil carbon and unsustainable soil loss due to erosion will depend on many factors such as soil type, slope, crop rotation and tillage practices.
Table 2. Stover residue (percent of ground covered) following stover harvest operations. | |||
---|---|---|---|
Site |
Harvested (h) |
Not harvested (nh) |
Difference (h-nh) |
A |
78% |
90% |
-12%* |
B |
73% |
92% |
-19%* |
C |
90% |
97% |
-7% |
Average |
82% |
95% |
-13%* |
*Difference is statistically significant (α=0.10).
Approximately 1–2 dry tons per acre of stover were removed from plots on average, and the resulting amounts of macronutrients (N, P2O5 and K2O) removed are summarized in Table 3. When averaged over years and locations, 10, 3 and 24 pounds of N, P2O5 and K2O, respectively, were removed per dry ton of stover. This is lower than the values (22.0, 8.2 and 32.0 pounds of N, P and K, respectively, per dry ton) cited by Warncke et al. in MSU Extension Bulletin E2904, “Michigan Corn Stover Project: Cattle, Storage and Bioenergy, MSU Extension bulletin E3354