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

Action: Provide nest boxes for birds Farmland Conservation

Key messages

Read our guidance on Key messages before continuing


Supporting evidence from individual studies


A before and after study of Eurasian kestrel Falco tinnunculus from 1959 to 1965 in the Oostelijk Flevoland in the Netherlands (Cavé 1968) found that there was an increase from 20 breeding pairs to 109 clutches following nest box installation in the study area (natural vegetation, plantation and 20% crops). In 1959, there were 25 nest boxes and approximately 20 breeding kestrel pairs, of which 11 used a nest box. In 1960, once the 243 boxes had been installed, 109 clutches were found in the three blocks (two natural vegetation, one cultivated); only a few were outside the boxes. In 1960-1965, there were 16-62 kestrel clutches in the cultivated block (80% plantation, 20% common crops; 117 boxes), there were also 0-12 long-eared owl Asio otus clutches. In each block, 81 kestrel nest boxes (50 x 30 x 30 cm) on 2 m poles were placed in nine rows of nine (330 m apart). An additional 36 nest boxes were placed in one half of the cultivated block in the winter of 1961-1962.



A controlled study in mixed farmland in northeast Scotland in 1971 (Yom-Tov 1974) found that carrion crows Corvus corone did not nest in artificial trees, irrespective of whether they were provided with supplementary food or not. In one experiment, a line of 15 artificial trees (3-6 m branches tied to fence posts and provided with an old crow’s nest) were set up, approximately 70 m apart. Two pairs of crows established territories, but neither attempted to breed. A second experiment provided a single artificial tree in two occupied territories, 70 m from the tree used by the resident pair. Neither artificial tree was used, as the resident pairs successfully defended their territories.


A replicated study of 48 common starling Sturnus vulgaris nest boxes at two pasture sites in Sweden (Svensson 1991) found that compared to boxes at 1.5 m or 3 m above ground, those at 4.5 m had significantly higher occupancy (100% vs 75%), numbers of eggs (94 vs 55-57) and numbers of hatched young (86 vs 46-53).  Mean date for first egg was also 2.6 days earlier in the highest boxes. Although there was also a greater number of fledged young in the highest boxes (27 vs 11-19), the average number fledged did not differ significantly between heights (3.4 vs 2.3). Wooden nest boxes (50 mm diameter entrance) were put up at the three heights on eight trees at each site. Boxes were inspected throughout the breeding season.



A replicated study in agricultural sites in southern Finland (Valkama & Korpimaki 1999) found that Eurasian kestrel Falco tinnunculus occupied 18-22% of 161 nest boxes between 1985 and 1995, with no differences between small, intermediate and large boxes. Boxes sheltered from prevailing weather were more likely to be occupied than exposed boxes (25% of 80 sheltered boxes used vs 17% of 81 exposed boxes). There were no significant differences in clutch size or number of fledglings produced between nest box types and orientations, with success related to laying date and vole Microtus spp. abundance. Occupied boxes were, on average, further from forest edges, roads and inhabited houses, and closer to grassy ditches than unoccupied boxes. Boxes were 25 × 27.5 × 25 cm, with a 12.5 x 25 cm entrance (small); 34 × 35 × 20 cm, with a 12 x 34 cm entrance (intermediate); or 33.5 × 45 × 30 cm, with a 12 x 33.5 cm hole (large).



A replicated study of common starling Sturnus vulgaris nest boxes in southern Sweden (Smith & Bruun 2002) found that nest success within boxes was high and was related to the amount of pasture available. There were between 1 and 8 (average 4) breeding attempts initiated in each colony of eight boxes. Only 8% of 609 nests failed during laying or incubation and an additional 5% during nestling rearing. Breeding attempts and the proportion of hatchlings that fledged increased and nest failures decreased with an increase of pasture in the surrounding area. In 1994, 19 breeding colonies of a row of eight nest boxes on trees (5-10 m apart, 1.8 m above ground) were established. An additional 13 colonies were installed in 1996-1998. Nest boxes were visited every 1-2 days to record egg-laying, hatching and fledging.



A replicated, paired site study from March-August in 2000-2003 in 20 paired nest box groups (10 placed along wetland edges and 10 in farmland) in Rutland, England (Field & Anderson 2004) found that tree sparrow Passer montanus showed a significant preference for nest boxes in wetland habitat, compared to those in farmland sites (eight wetland nest boxes colonized vs no farmland sites). Nest box groups consisted of five nest boxes placed 2-20 apart. Sunflower seeds were randomly provided to one nest box group within each pair.




A replicated trial in arable farming landscapes in Norfolk, England, in the summers of 1997-2001 (Browne 2006) found that tits Parus spp. nested in a higher proportion of hanging woodcrete boxes (38% of 48 boxes occupied), compared to tree-mounted woodcrete boxes (25% of 48) or thick and thin wooden boxes (20% and 16% of 48 boxes respectively). Patterns were the same for great tit Parus major, blue tit P. (Cyanistes) caeruleus and all species combined (also including coal tit P. (Periparus) ater and marsh tit P. (Poecile) palustris), although a higher proportion of great tits used woodcrete boxes (91% of great tits vs 47% of blue tits). Clutch size, brood size and number of young fledged by blue tits and great tits did not differ significantly between box types. Woodcrete boxes were either attached to a tree trunk (18 cm high, base 18 cm diameter) or free-hanging (19 cm high, base 11 cm diameter). Wooden boxes were 16.5 x 15 x 19.5 cm, and of either 1.9 cm or 2.4 cm thick wood. All designs had a 3.2 cm diameter entrance. Another trial found that a higher proportion of tit Parus spp. nests were in 50 green nest boxes (72% of 41 nests) than in 50 brown boxes (28%), and in 50 boxes with circular entrances (68%) compared to those with a wedge-shaped entrance (32%).



A trial from 2003 to 2005 on a single farm, Rawcliffe Bridge, East Yorkshire, UK (Bryson et al. 2007) found that nest boxes were 54%, 50% and 68% occupied in 2003, 2004 and 2005 respectively. In 2003, all five boxes designed for tree sparrow Passer montanus were occupied. In 2005, 20 tree sparrow boxes (70% of the 28 provided) were occupied. The number of breeding tree sparrows on the farm increased from 6 to 20 pairs between 2003 and 2005. In the years 2003, 2004 and 2005, 32, 60 and 84 bird nest boxes were put up, including some designed for tree sparrows. They were inspected in February each year. Birds on the farm were monitored five times each year from 2003 to 2005, by walking the field boundaries. The number of breeding pairs/ha was estimated from clusters of sightings.



A replicated study in 1988-2000 in Noord-Brabant, the Netherlands (Potters 2009) found that stock pigeon Columba oenas used nest boxes provided in mixed agricultural habitats, laying in total an average of 118 eggs laid/year with 52% hatching and 84% of chicks fledging (an average of 52 chicks/year). Boxes were 20 x 20 x 50 cm, with an 8 x 8 cm entrance hole and placed 3-5 m above the ground in trees, 20-30 m apart. Jackdaw Corvus monedula also used the nest boxes, but were removed from 1995 onwards.



A replicated study from 2002 to 2008 of 625 nest boxes inside agricultural shacks and buildings in Valais, Switzerland (Zingg et al. 2010) found that 5% were occupied by Eurasian wryneck Jynx torquilla in 2008. Of the 269 monitored locations (2-3 boxes/location), 32 (12%) were occupied by a wryneck in 2008; 23 of those locations had a wryneck nest box. Within the occupied locations, 19 wryneck broods occurred in one of the 56 available Eurasian hoopoe Upupa epops nest boxes and 14 occurred in one of the 22 available wryneck nest boxes. Locations that had been occupied in the past had a higher probability of occupancy. The presence of hoopoes had no influence on the nest box choice. Wryneck nest boxes had no effect on reproductive output, however, in general, nestlings from broods in wryneck nest boxes had a higher body mass than those in hoopoe boxes (27 vs 25 g). The wryneck population inhabiting the hoopoe nest boxes declined from 72 broods in 2002 to 34 broods in 2007, potentially due to competition with the hoopoe population (1998: 20 broods, 2007: 160 broods). The study site was largely of fruit plantations, vineyards and vegetable cultures. A pair of hoopoe boxes were installed at each location from 1998 to 2003 and a further 135 wryneck boxes were installed at half of the locations in 2008.

Referenced papers

Please cite as:

Dicks, L.V., Ashpole, J.E., Dänhardt, J., James, K., Jönsson, A., Randall, N., Showler, D.A., Smith, R.K., Turpie, S., Williams D.R. & Sutherland, W.J. (2019) Farmland Conservation Pages 291-330 in: W.J. Sutherland, L.V. Dicks, N. Ockendon, S.O. Petrovan & R.K. Smith (eds) What Works in Conservation 2019. Open Book Publishers, Cambridge, UK.