EAST LANSING, Mich. — In areas with high white-tailed deer densities, damage to field crops through feeding, trampling and bedding can reduce harvestable yields. Research by Fritzell et al. in 1995 has suggested that crop losses to deer greater than 10-15 percent of the total crop or $20 per acre are viewed as significant and requiring remedy by Michigan producers and hunters. Northeast Lower Michigan has been identified as an area of special concern for white-tailed deer management due to deer population densities well-above Department of Natural Resources goals and the high prevalence of crop depredation, according to the Michigan Department of Natural Resources (MDNR) in 2010. Damage to soybeans, in particular, has become problematic for growers in the region as production of the crop has expanded to approximately 25,000 acres over the last 25 years.
Soybeans are a favorite food of white-tailed deer, meeting their dietary requirements all season long, according to Colligan in 2011). Substantial defoliation by deer (67-100 percent) prior to the V6 growth stage can significantly reduce soybean yields, according to research by Garrison and Lewis in 1987.
On the other hand, according to research by Rogerson et al. in 2014, mild to moderate damage has been shown to increase soybean yield in some cases by stimulating additional branching and pod set. Deer damage is more likely in small fields with a high proportion of forested edge, according to Braun in a 1996 study.
Research conducted in the region by Michigan State University Extension during 2015 recorded early-season defoliation in soybeans ranging from 0 to 87.5 percent. Deer browsing significantly reduced soybean yields that year. Yield loss at the field scale ranged from 0 to 100 percent, averaging 10 percent or 3.89 bushels per acre.
Chemical repellents are one tool available to growers for controlling wildlife crop damage. Putrescent egg solids have been used successfully as a deer repellent and included as an active ingredient in several commercial repellent products. Decomposing eggs emit a sulfurous odor that is thought to mimic the smell of predator urine and thus discourage deer from browsing on treated foliage. DeerPro Spring & Summer manufactured by Great Oak Inc. of Redding, CT is another commercial repellent product containing 2.6 percent putrescent egg solids.
In a 2016 study by Clemson University, DeerPro significantly reduced deer damage to soybeans relative to an untreated control and other repellent products. In 2017, Presque Isle County MSU Extension trialed a new formulation of DeerPro Spring & Summer with capsaicin added as a means of reducing deer depredation of soybean.
Materials and methods
Two separate commercial soybean fields in Presque Isle County were selected for the trial based on landscape factors known to increase the likelihood of deer damage (forested field edges), and a history of significant deer damage as reported by cooperating growers. Site one was intended as a trial run to gain familiarity with the product, whereas site two included more robust experimental design to increase the probability of detecting treatment differences. In the end, weather and other production constraints resulted in methodological differences between the two sites that have allowed us to speculate about best practices for the use of DeerPro Spring & Summer in soybeans.
Site one was located in Metz Township on a 50-acre soybean field bordered on 3.5 sides by forestland and adjacent to large tracts of woodland maintained for wildlife and hunting. At this site, MG 1.7 soybeans were drilled in 7.5-inch rows at 250,000 seeds per acre on May 21, 2017. Plots 90 feet wide by 200 feet long were established running parallel to the west field edge shortly after planting. DeerPro and untreated control treatments were assigned to every other plot for convenience and replicated three times for a total of six plots.
DeerPro Spring & Summer was applied at a rate of 1.25 gal per acre carried in 8.75 gal of water (10 gallon per acre total rate) using the grower’s self-propelled sprayer on May 24 at emergence (VE), and again on June 24 at the third trifoliate (V3) stage. Both repellent applications were made immediately following, but separate from, glyphosate applications for weed control. One 5-foot by 5-foot exclusion cage constructed of steel fence posts and Tenax C flex plastic mesh fencing was randomly placed in each plot to measure soybean development and yield potential with zero deer pressure.
Site two was located in Case Township on a 53-acre soybean field bordered on four sides by forest and adjacent to private property where deer are fed. At this site, MG 0.8 soybeans were planted in 30-inch rows at 175,000 seeds per acre on June 5, 2017. Plots 90 feet wide by 300 feet long were established running perpendicular to the east field edge shortly after emergence. DeerPro and untreated control treatments were randomly assigned and replicated four times for a total of eight plots arranged in a completely randomized design.
DeerPro Spring & Summer was applied at the same rate as site one using the grower’s self-propelled sprayer on June 22 at the first trifoliate (V1) stage, and again on July 18 at the four trifoliate (V4) stage. Both repellent applications were made immediately following, but separate from, glyphosate applications for weed control. Two 4-foot by 4-foot exclusion cages constructed of steel fence posts and Tenax C flex plastic mesh fencing were placed in each plot at approximately 30 feet and 150 feet from the field edge to measure soybean development and yield potential with zero deer pressure.
Deer damage was monitored at both sites from emergence until the V6/R1 growth stage, after which damage is thought to be less detrimental to yield. Damage was measured using a few different methods in an effort to compare their power as predictors of yield loss. Our primary method was hand counting the number of plants damaged and percent defoliation relative to undamaged plants in 15-inch by 15-inch quadrats at four locations per plot every 10 or so days. Counting began on June 6 at site one, continuing for eight weeks, and began on June 22 at site two, continuing for five weeks. We also used an application developed by Oklahoma State University called Canopeo to measure soybean canopy cover beginning on July 10.
Lastly, we hired Mobile Drone Systems LLC. and Skymatics Inc. of Alberta, California, to collect remote estimates of green plant area versus soil and soybean canopy height (site two only) using a UAV at site one on July 15 at the V5 stage and site two on July 14 at the V3 stage. At crop maturity, two methods were used to measure soybean yield on Oct. 15 at site one and Oct. 20 at site two. The first method was to harvest soybeans within each exclusion cage. Beans were hand-harvested from the cages, threshed using a stationary plot thresher and weighed.
After the exclusion samples were harvested and cages removed, a strip was harvested from each plot using the cooperating grower’s combine (60-feet by 200-feet at site one, 80-feet by 300-feet at site two) and weighed using a weigh wagon to determine average plot yield. All yield data was adjusted for grain moisture at harvest.
Results and discussion
At site one, the DeerPro Spring & Summer treatment significantly reduced defoliation at 27, 43 and 49 days after emergence and also significantly reduced yield loss due to deer feeding by 10.42 bushels per acre (30 percent) relative to the untreated control. However, the DeerPro treatment had no significant effect on canopy cover as measured by Canopeo, or on green plant fraction as measured by the UAV at the V5 growth stage (Tables 1 and 2).
At site two, the DeerPro treatment significantly reduced defoliation only at 49 days after emergence (following the second application), and did not significantly reduce yield loss due to deer feeding. The DeerPro treatment did numerically reduce the average difference in canopy cover between exclosures and exposed areas by 14.5 percent and average difference in height by 3.2 centimeters (13.6 percent), but these differences were not quite statistically significant at the 0.10 level. The DeerPro treatment had no significant effect on green plant fraction as measured by the UAV at the V3 growth stage (Tables 3 and 4).
Across all methods and timings, hand counting of percent defoliation at 40-43 days after emergence was the most robust predictor of yield loss due to deer feeding. Hand counts of percent defoliation were also decent predictors of yield loss when conducted at other timings, but not as good as when taken at 40-43 days after emergence. Other less direct measures of deer damage (canopy cover, green plant fraction, canopy height) were poor predictors of yield loss in our experiment, which may be due to their inability to fully capture defoliation that does not reduce canopy cover, and the general lack of correlation between soybean height and yield.
The different outcomes observed between sites in our trial may suggest scenarios where DeerPro is more likely to be effective or best management practices for the use of DeerPro Spring & Summer in soybeans. At the most fundamental level, there were differences between our sites in terms of deer pressure. While we did not actually measure deer numbers or time spent by deer in each field, percent defoliation averaged 19.5 percent at site one versus 37.1 percent at site two 43 days after emergence.
Yield potential independent of deer browsing was also very different at our two locations. Average yield inside exclosures at site one was 27 bushels per acre versus 42 bushels per acre at site two. Deer may be more or less attracted to food sources based on their abundance and quality, and the high deer pressure at site two may have simply overwhelmed any treatment effect.
Managing the two trial locations also differed dramatically in terms of plant dates, seeding rates, row spacing, application timing, etc. For example, site one was planted and emerged approximately two weeks earlier than site two. This may have made site one an attractive early food source for deer, but likely not any more attractive than other soybeans emerging at a similar pace nearby.
Conversely, site two was planted late for our region in an area where few other soybeans are grown. Thus, this field remained lush and palatable later into the summer while alternative food sources in the area were either nonexistent, or rapidly declining in palatability with increasing maturity.
Furthermore, site one was drilled in narrow rows at a much higher planting population (250,000 seeds per acre) than site two, which was planted in wide rows at only 175,000 seeds per acre. Narrow row spacings and higher seeding rates may increase the capacity of soybeans to recover from early season defoliation relative to wide rows and lower seeding rates.
Application timing also differed between our two sites, which may be the most important factor in our results. DeerPro was applied at site one during soybean emergence before any deer damage was detected and again at V3. This can be contrasted with site two where DeerPro was applied at V1 and V4. At site two, damage had already begun with defoliation averaging 15.3 percent by the time our first application was made. Several references suggest that wildlife repellents are much more effective when applied prior to the initiation of damage (i.e., site one).
Finally, an interesting point to note: Our data suggests that deer depredation can cause significant yield loss in soybean at levels well below the threshold of 67 percent defoliation prior to V6 suggested in the literature. Yield reductions greater than 25 percent were observed with only approximately 20 percent defoliation at 43 days after emergence in three plots. While our defoliation measure was not a perfect representation of biomass removal due to the difficulty of accounting for regrowth after damage, a large amount of defoliation was not required to significantly reduce soybean yield.
|Table 1. Depredation differences in/out of exclosures 40-43 days after emergence by treatment at site one, mean separation at the 0.10 level.|
|Treatment||Defoliation (%)||Canopy cover (%)||Plant fraction (%)|
|DeerPro||11.43 b||-1.90 a||-22.07 a|
|Control||27.60 a||-16.17 a||-23.10 a|
|Table 2. Yield differences in/out of exclosures by treatment at site one, mean separation at the 0.10 level.|
|Treatment||Yield (bu/a)||Yield (%)|
|DeerPro||-2.84 b||-11.87 a|
|Control||-13.14 a||-41.83 a|
|Table 3. Depredation differences in/out of exclosures 40-43 days after emergence by treatment at site two, mean separation at the 0.10 level.|
|Treatment||Defoliation (%)||Canopy cover (%)||Plant fraction (%)||Height (cm)|
|DeerPro||41.28 a||-11.25 a||-13.50 a||-4.43 a|
|Control||32.90 a||-25.75 a||-17.98 a||-7.63 a|
|Table 4. Yield differences in/out of exclosures by Treatment at site two, mean separation at the 0.10 level.|
|Treatment||Yield (bu/a)||Yield (%)|
|DeerPro||-22.15 a||-54.38 a|
|Control||-23.03 a||-52.90 a|
- Braun, K. 1996. Ecological factors influencing white-tailed deer damage to agricultural crops in northern lower Michigan. M.S. Thesis. MSU Department of Fisheries and Wildlife
- Campa, H., III, S. R. Winterstein, R. B. Peyton, G. Dudderar, and L. A. Leefers. 1997. An evaluation of a multidisciplinary problem: Ecological and sociological factors influencing white-tailed deer damage to agricultural crops in Michigan. Transactions of the North American Wildlife and Natural Resource Conference 62:431-440.
- Colligan, G.M., Bowman, J.L., Rogerson, J.E. and B.L. Vasilas. 2011. Factors affecting white-tailed deer browsing rates on early growth stages of soybean crops. Human–Wildlife Interactions 5(2):321–332.
- Conover, M. R., and D.J. Decker. 1991. Wildlife damage to crops; perceptions of agricultural and wildlife professionals in 1957 and 1987. Wildl. Soc. Bull. 19:46-52.
- Fritzell, P.A., Minnis, D.L. and R.B. Peyton. 1995. A comparison of deer hunter and farmer attitudes about crop damage abatement in Michigan: Messages for hunters, farmers and managers. Proceedings of the Seventh Eastern Wildlife Damage Management Conference.
- Fritzell, P., Dudderar, G. and R.B. Peyton. 1997. An Evaluation of Farmer Applications of Deer Damage Controls. Proceedings of the Eighth Eastern Wildlife Damage Management Conference
- Garrison, R.L. and J.C. Lewis. 1987. Effects of Browsing by White-Tailed Deer on Yields of Soybeans. Wildlife Society Bulletin,Vol. 15, No. 4. pp. 555-559
- MDNR. 2010. Michigan Deer Management Plan. Michigan Department of Natural Resources and Environment. Wildlife Division Report No. 3512
- Rogerson, J., Bowman, J., Tymkiw, E., Colligan, G. and B. Vasilas. 2014. The Impacts of White-Tailed Deer Browsing and Distance From the Forest Edge on Soybean Yield. Wildlife Society Bulletin 38(3):473–479.
- Trent, A., Nolte, D. and K. Wagner. 2001. Comparison of Commercial Deer Repellents. USDA National Wildlife Research Center – Staff Publications #572.
The authors would like to thank Great Oak Inc. for supporting this research, and the Brandt family and Tyler Tollini for their time and effort in conducting the trial.
— James DeDecker and Christian Tollini, Michigan State University Extension
For more news from Michigan, click here.