Use loud noises to deter crop damage (e.g. banger sticks, drums, tins, iron sheets) by mammals to reduce human-wildlife conflict

How is the evidence assessed?
  • Effectiveness
  • Certainty
  • Harms

Study locations

Key messages

  • Ten studies evaluated the effects of using loud noises to deter crop damage by mammals to reduce human-wildlife conflict. Three studies were in the USA, two were in Zimbabwe and Kenya and one each was in the UK, Namibia, and India.





  • Human-wildlife conflict (10 studies): Five of six studies (including two controlled, one replicated and two before-and-after studies), in the USA, Namibia, Kenya and India, found that loud noises activated when an animal was in the vicinity reduced or partially reduced crop damage or crop visits by white-tailed deer, black-tailed deer (when combined with using electric shock collars) and elephants. The other study found that using loud noises (along with chili fences and chili smoke) did not reduce crop-raiding by African elephants. Three studies (including two controlled studies), in the UK and the USA, found that regularly sounding loud noises did not repel European rabbits or white-tailed deer. Two replicated studies, in Zimbabwe, found that, from among a range of deterrents, African elephants were repelled faster from crop fields when scared by firecrackers or by a combination of deterrents that included drums.

About key messages

Key messages provide a descriptive index to studies we have found that test this intervention.

Studies are not directly comparable or of equal value. When making decisions based on this evidence, you should consider factors such as study size, study design, reported metrics and relevance of the study to your situation, rather than simply counting the number of studies that support a particular interpretation.

Supporting evidence from individual studies

  1. A before-and-after study in 1984 on grassland in Surrey, UK (Wilson & McKillop 1986) found that an acoustic scaring device did not deter European rabbits Oryctolagus cuniculus from consuming bait. Bait consumption after the device was activated (2–361 g/bait pile/day), did not differ from that before the device was activated (7–368 g/bait pile/day). Five wild, adult rabbits were placed in a 50 × 40-m grass enclosure, with wooden hutches at one end. The opposite end housed the scaring device and 400-g piles of chopped carrots at 3, 6, 9, 12 and 15 m from the device. The device emitted 5-s bursts of rapidly pulsed sound, separated by 4-s silences. Bait was deposited on four days/week. Remaining carrots were removed and weighed to establish quantity consumed. Similar bait, in rabbit-proof cages, was used to correct weights for moisture changes. The enclosure contained sufficient grass to sustain rabbits without their need to eat carrots. The trial lasted four weeks, in March 1984, with the scaring device switched on midway through.

    Study and other actions tested
  2. A randomized, controlled, before-and-after study in 1994–1995 on a grassland site in Ohio, USA (Belant et al. 1996) found that motion-activated propane exploders temporarily reduced white-tailed deer Odocoileus virginianus visits but regularly firing exploders did not. There were fewer deer visits in the week following deployment of motion-activated exploders, in two out of three seasons (23–94 visits/week) compared to the pre-treatment period (159–313 visits/week). In spring/early-summer and late-summer, visit rates returned to pre-treatment levels after 2–6 weeks. In autumn, exploders did not reduce deer visits. Regularly firing exploders did not reduce deer visit rates compared to pre-treatment levels in any weeks studied and neither did non-functioning exploders. The experiment used different combinations of three out of six feeding sites, during 9 August–12 September 1994, 20 September–24 October 1994 and 27 April–12 July 1995. Each time, a two-week pre-treatment period preceded a 3–9-week treatment period. Feeding sites (>1 km apart) were semi-circular fences around whole kernel corn. Treatments were propane exploders firing eight times in two minutes when motion was detected, exploders firing every 8–10 minutes and non-functioning exploders. Deer visits were monitored with electronic detecting devices.

    Study and other actions tested
  3. A replicated study in 1993–1995 of farmland and grassland at 10 villages in East Caprivi, Namibia (O'Connell-Rodwell et al. 2000) found that car sirens connected to trip wires around crops were partially successful in reducing crop raiding by elephants Loxodonta africana. Sirens at three villages in the first year were all reported to have positive effects of reducing crop-raiding by elephants (actual crop-raiding frequencies not reported). In the second year, a positive effect of sirens was reported from one village, whilst elephants did not approach at three villages (so the system was untested) and at two further villages, the crop area was too large to protect using the system. In the third year, three villages reported positive effects whilst at a fourth, battery failure rendered the system ineffective. Sirens each protected 1–7 farms at 10 villages during one or two years of the trial. Each system comprised a car siren, a 12-V battery and a 10-s timer. Polyethylene cords were mounted on fences or trees to enclose fields. The siren activated for 10 s when the cord was pulled. Data were collated from questionnaire surveys in 1993–1995.

    Study and other actions tested
  4. A replicated study in 1995–1996 in crop fields at a site surrounded by savanna in Sebungwe, Zimbabwe (Osborn 2002) found that African elephants Loxodonta africana were repelled faster from agricultural fields by groups of people banging drums (alongside a range of other deterrents) than by one person making less noise. Specific effects of banging on drums cannot be separated from those of other scaring tactics. Elephants were repelled faster when scared by people with drums, dogs Canis lupus familiaris, whips and large fires (4 minutes) or with drums, dogs, slingshots and burning sticks (10 minutes) than by one person sometimes with a dog and chasing elephants while banging on tins and yelling (14 minutes). When scared by actions that included drums, elephants charged at defenders 12 times out of 26 trials, though only charged two out of nine times when scared by a single person without drums. Elephants raiding crops were scared 15 times by 4–7 people with drums, dogs, whips and large fires, 11 times by 2–3 people with drums, dogs, slingshots, and burning sticks and 15 times by one person (sometimes with a dog, and sometimes hitting tins and yelling to deter elephants). Behavioural responses were monitored through a monocular. Distance between elephants and farmers was 20–40 m. Tests were conducted between 18:30 and 06:30 h. The number of fields was not specified.

    Study and other actions tested
  5. A replicated study in 2001 of arable land in seven villages in Guruve District, Zimbabwe (Osborn & Parker 2002) found that using loud noises, by throwing firecrackers at crop-raiding elephants Loxodonta africana, repelled them faster than did traditional deterrents such as beating drums and throwing rocks. Elephants left faster when firecrackers were activated (average 6 minutes) than they did when traditional repellent methods alone were used (average 65 minutes). Seven villages were studied. At three villages, on 35 occasions, farmers threw locally made firecrackers at elephants that were attempting to raid crops. On 27 occasions, farmers at four villages used traditional methods to ward off elephants that attempted to raid crops, namely banging drums and throwing rocks with catapults. The study was conducted from 1 January to 30 June 2001 and data were collected by a team of observers.

    Study and other actions tested
  6. A replicated, controlled study in two pastures in Washington, USA (Nolte et al. 2003) found that playing loud noise, along with using shock collars, reduced damage by black-tailed deer Odocoileus hemionus to tree seedlings. The loud noise and electric shock were part of the same treatment, so their relative effects could not be separated. In areas where playing of loud noise was triggered, damage to tree seedlings was lower (0–1 bites) than in areas where loud noises were not triggered (0–25 bites). Three deer, fitted with shock collars, were placed in each of two 1.5-ha pastures. Within each pasture, four 20 × 20 m plots were established. In each plot, 16 red cedar Thuja plicata seedlings were planted at 1-m intervals. When deer entered two of the plots, a loud noise was played through a speaker and deer received an electric shock. When they entered the other two plots, no noise was played and they received no shock. Deer activity was measured by counting the number of bites taken from seedlings over a 21-day period.

    Study and other actions tested
  7. A replicated, paired sites, controlled study in 1999 of corn fields at two sites in Nebraska, USA (Gilsdorf et al. 2004) found that loud noises from propane exploders did not reduce visits to crops by white-tailed deer Odocoileus virginianus. The number of deer visits/km of field boundary was similar in fields protected by propane exploders (31–36/day) and unprotected fields (40–56/day). Similarly, there were no significant difference between fields before devices operated (exploders: 17 visits/km/day; unprotected: 21 visits/km/day) or after (exploders: 37 visits/km/day; unprotected: 53 visits/km/day). Four groups of fields (0.5–2.5 km apart, separated by woodland) were studied at each of two sites. At each site, one field had propane exploders (two/field) and one was unguarded. Propane exploders fired at 15-minute intervals. They operated at night, from when corn crops became susceptible to damage (13 July 1999 at one site and 25 July 1999 at the second site), for 18 days. Deer activity was assessed by counting tracks twice while devices operated and once each in ≤2 weeks before and after this time.

    Study and other actions tested
  8. A before-and-after and site comparison study in 2003–2004 of two farming areas in Laikipia, Kenya (Graham & Ochieng 2008) found that using loud noises, along with chili fences and chili smoke, reduced raiding and crop damage by African elephants Loxodonta africana. The study does not distinguish between the effects of loud noises and chilli deterrents. After farmers began using loud noises, along with chili fences and smoke, the total number of crop-raiding incidents (26) and the average area of crop damage (375 m2/incident) was lower than before deterrents were used (92 incidents; 585 m2/incident). However, the difference was not tested for statistical significance. At a control site without deterrents, crop-raiding increased (total 17–166 incidents) as did crop damage (average 328 m2–421 m2/incident) during the same time period. A group of farmers within a 0.03-km2 area were provided with training and materials to deter crop-raiding elephants. Deterrents included loud noises (bangers, banger sticks, cow bells), chili fences (rope and cloth fences with chili and engine grease applied) and chili smoke (chili and dung briquettes burned at night). Some farmers also used watchtowers and torches. A second control area, of equal size and within 1 km, used no deterrents. Crop-raiding incidents and crop damage were recorded in each of the two areas before (June–December 2003) and after (June–December 2004) deterrents were introduced.

    Study and other actions tested
  9. A replicated, before-and-after and site comparison study in 2004–2005 at 40 farms in Laikipia, Kenya (Graham & Ochieng 2008) found that using loud noises, along with chili fences and chili smoke, did not result in an overall reduction in crop-raiding by African elephants Loxodonta africana. The study does not distinguish between the effects of chilli deterrents and loud noises. After farmers began using loud noises, along with chili fences and chili smoke, the average number of crop-raiding incidents across all farms (2) was similar to before deterrents were used (2.5). At 10 control farms without deterrents, crop-raiding decreased (from an average of three incidents to one) during the same time period. Ten farmers in each of two areas were provided with training and materials to deter crop-raiding elephants. Deterrents included loud noises (bangers, banger sticks, cow bells), chili fences (rope and cloth fences with chili and engine grease applied) and chili smoke (chili and dung briquettes burned at night). Some farmers also used watchtowers and torches. Uptake of deterrent types varied between farms (see original paper for details). Ten control farms within each of the two areas used no deterrents. Crop-raiding incidents were recorded at all 40 farms before (February–November 2004) and after (February–November 2005) deterrents were introduced.

    Study and other actions tested
  10. A study in 2006–2009 in two areas of Assam, India (Davies et al. 2011) found that using loud noises to scare Asian elephants Elephas maximus reduced the probability of elephants damaging crops. The chance of crop damage occurring was lower when noise was used to deter elephants compared to a range of other interventions or to no intervention (results presented as statistic model coefficients). Only fences and spotlights reduced crop raiding to a greater extent. Within two study areas, 33 community members, trained as monitors, recorded 1,761 crop-raiding incidents, from 1 March 2006 to 28 February 2009. A range of deterrent methods was used, singly or in combination, including noise (shouting, crackers or drums), chili smoke (from burning dried chilies, tobacco, and straw), spotlights, two-strand electric fences, chili fencing (engine grease and ground chili paste, on a jute or coconut rope), elephant drives (repelling wild elephants using domesticated elephants) and fire.

    Study and other actions tested
Please cite as:

Littlewood, N.A., Rocha, R., Smith, R.K., Martin, P.A., Lockhart, S.L., Schoonover, R.F., Wilman, E., Bladon, A.J., Sainsbury, K.A., Pimm S. and Sutherland, W.J. (2020) Terrestrial Mammal Conservation: Global Evidence for the Effects of Interventions for terrestrial mammals excluding bats and primates. Synopses of Conservation Evidence Series. University of Cambridge, Cambridge, UK.

Where has this evidence come from?

List of journals searched by synopsis

All the journals searched for all synopses

Terrestrial Mammal Conservation

This Action forms part of the Action Synopsis:

Terrestrial Mammal Conservation
Terrestrial Mammal Conservation

Terrestrial Mammal Conservation - Published 2020

Terrestrial Mammal Conservation

What Works 2021 cover

What Works in Conservation

What Works in Conservation provides expert assessments of the effectiveness of actions, based on summarised evidence, in synopses. Subjects covered so far include amphibians, birds, mammals, forests, peatland and control of freshwater invasive species. More are in progress.

More about What Works in Conservation

Download free PDF or purchase
The Conservation Evidence Journal

The Conservation Evidence Journal

An online, free to publish in, open-access journal publishing results from research and projects that test the effectiveness of conservation actions.

Read the latest volume: Volume 20

Go to the CE Journal

Discover more on our blog

Our blog contains the latest news and updates from the Conservation Evidence team, the Conservation Evidence Journal, and our global partners in evidence-based conservation.

Who uses Conservation Evidence?

Meet some of the evidence champions

Endangered Landscape Programme Red List Champion - Arc Kent Wildlife Trust The Rufford Foundation Save the Frogs - Ghana Bern wood Supporting Conservation Leaders National Biodiversity Network Sustainability Dashboard Frog Life The international journey of Conservation - Oryx British trust for ornithology Cool Farm Alliance UNEP AWFA Bat Conservation InternationalPeople trust for endangered speciesVincet Wildlife Trust