Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Individual study: A frightening device for deterring white-tailed deer Odocoileus virginianus use of cattle feeders in Montmorency County, Michigan, USA

Published source details

Seward N.W., Phillips G.E. Duquette J.F. & Vercauteren K.C. (2007) A frightening device for deterring deer use of cattle feeders. Journal of Wildlife Management, 71, 271-276


The presence of bovine tuberculosis (TB) in cattle can negatively impact the cattle industry. Deer most likely contract the disease from infected cattle and now act as the main reservoir for re-infecting cattle herds in some US states, including Michigan. Management strategies to reduce TB transmission have included increasing hunter harvest to reduce deer densities, deer culling, removal of infected cattle and instating non-lethal means (e.g. frightening devices, dogs and fencing) to reduce cattle-deer interactions. Since bovine TB can be transmitted through shared feed, a frightening device was developed to deter deer from using cattle feeders.

Study area: The study was conducted on two farms in Montmorency County in northeastern lower Michigan (45º01'N, 84º07'W). The first phase of the study took placed on a privately owned 34.2 ha deer enclosure containing 105 white-tailed deer (equivalent to 307deer/km²); there was little natural vegetation available for forage in the enclosure, as indicated by the browse-line throughout the understorey. Vegetation in the enclosure consisted of approximately 90% forest cover and 10% open grassland. The second phase of the study took place on a 50 ha cattle farm with 35 cattle to assess the effectiveness of a deer-resistant cattle feeder.

Deer frightening device: The device was constructed using an 18-gauge high-tensile tubing round-bale feeder as the frame to suspend a wooden platform and cabinet that housed the electronics and gear motor to power the system. The gear motor rotated a 3.4 m aluminium bar with two 1.6 m long polyvinyl chloride (PVC) conduit arms that hung down from steel chains to 45 cm above the ground.

Sensors detected approaching deer, which triggered the rotating arms which struck the deer if within a metre of the feeder. The force of the impact delivered by PVC arms was minimal and designed to startle deer, not to hurt them.

Cattle feeders were installed at six feeding sites well distributed within the deer enclosure, three of which were fitted with the frightening device. Feeders were filled with corn Zea mays and other fodder known to be attractive to the deer. Animal-activated infra-red cameras were used to collect data on deer use of the six feeders. Deer use of all the feeders was monitored continuously during a 4 week pre-treatment period (10 February-10 March), a developmental period (13 March-4 April) when design flaws were identified. The final prototype device was implemented during the treatment period (13 May-23 June).

Second phase experiments: Subsequent experiments were carried out at a cattle farm to determine whether deer-resistant feeders would affect cattle behaviour and assess the durability of the device (13 July–12 August). Cattle were rotated among the 10, 5 ha grassland pastures at the 50 ha farm and animal-activated infra-red cameras were used to collect data. To simulate how the feeder might be practically used by managers, the frightening device was only activated at night (when contact with deer was most likely).

Deer use was found to be similar at protected and unprotected feeders during the pre-treatment period, but was lower at protected feeders during the treatment period (after improvements made to the design). Feeders fitted with the deterrent device were 100% effective during the first two weeks when activated. However, its effectiveness dropped over time, averaging 94% during the first five weeks and 61% in the final (sixth) week.

The device efficiency was compromised by the inability of the infra-red sensors to reset quickly enough in order to reactivate the deterrent device (due to continuous detection of animal movement). This therefore allowed habituated deer to feed unhindered. Furthermore, other species, such as wild turkey Meleagris gallopovo and grey squirrel Sciurus carolinensis also activated the frightening device, which further depleted battery power.

Although the frightening device may initially deter cattle, evidence showed that they continued to use the feeders and ignored the swing arms.

In terms of durability in the presence of cattle, the feeder was found to be inoperative 17 days out of 30 days during the cattle trial. Problems included sensor or battery malfunction, hanging PVC arm caught in the feeder, and broken support arms.

Conclusions: Unlike cattle, deer were scared from feeders when struck by the PVC arms of the frightening device. When the deterrent device operated, it was over 90% successful in deterring deer; however, some modifications to the design are required to eradicate problems related to battery power and infra-red sensors. The deterrent device would also have to be strengthened and improved to prevent damage by cattle.

Note: If using or referring to this published study, please read and quote the original paper.