Associations between human non-motorized recreational activity on nest box occupation, exploratory behaviour, and breeding success in a passerine bird

Anthropomorphic activities have a large impact on ecosystems in many ways, one of which is how animals behave. Non-motorised nature recreation is a popular human activity of which the impacts on nature are largely unknown. These activities, which include hiking, biking, pet walking and horseback riding, tend to increase during the commencement of the breeding activity for most passerine forest birds in temperate zones. We here investigated whether variation in recreational activity associates with patterns of nest box occupation and reproductive success in a long-term study of personality-typed great tits (Parus major). We measured human disturbance in the area by recording the frequency of non-motorised recreational activities by observations. We were particularly interested in the relationship between disturbance levels and nest box occupancy as well as the relationship between disturbance levels of occupied nest boxes and exploratory scores of the great tits that occupied them. We also investigated whether reproductive characteristics such as fledging success, clutch size, chick weight and tarsus length varied with disturbance levels at occupied nest boxes. We did not find a relation between nest box occupation and disturbance. Habitat quality rather than disturbance explained the nest occupation. More exploratory individuals occupied boxes in less disturbed areas, independent of habitat quality. Fitness decreased with increasing disturbance independent on habitat quality. Chicks were heavier and had longer tarsi, and clutch sizes were bigger in less disturbed areas. In conclusion, we found breeding site choice of great tits to be independent on human activity, although there are clear fitness effects of human disturbance.

Methods
Data Collection

The disturbance data is an observational data. To measure the disturbance caused by recreational activities, we counted the number of people we encountered on the 12 path junctions in the study area. The activities that were considered recreational activities were hiking, walking, cycling, pet walking and horseback riding. On each junction, the measurements were taken the same number of times in two separate time periods that we pre-determined morning (9-12h) and afternoon (12-15h). A kernel interpolation was performed using ArcGis Pro version 3.1.0 to estimate the disturbance levels of each nest box in the study area. Kernel values were calculated by taking into account the number of visitors on the paths and the distance of nest box to all paths around it.

The nest occupancy data was extracted from the nest box breeding occupancy data that were collected during breeding seasons (April-June) from 2012 to 2021. In order to determine the individuals that occupie the nest aduly birds were caught inside the nest box using spring traps on day 7 after hatching.

Reproductive measures: 1) laydate 2) clutch size 3) chick biometric measures (weight and tarsus length) and 4) the number of fledged chicks, were collected during breeding seasons (April-June) from 2012 to 2021. To record these measures, bi-weekly checks were conducted on nest boxes in order to determine the state of the nest building from the end of March onwards. When nest building occurred, nest boxes were checked every other day in order to determine the exact laying date, clutch size and start of incubation. The exact hatch date was determined by checking the broods daily around the expected hatch date. On day 14 after hatching, we measured tarsus length (0.1mm) and body mass (0.1gr). To determine the number of fledged chicks, we checked nest boxes 22 days after hatching.

Habitat quality: During the 2019 breeding season, we assessed the quality of the territory around each nest box (50m radius). Based on the two most abundant tree species within the territory we assigned each nest box to be of “high” or “low” quality. A territory was considered to be of ‘high’ quality when the two most abundant tree species were deciduous (pedunculate oak, Quercus robur; beech, Fagus sylvatica; red oak, Q. rubra; birch, Betula pendula; larch, Larix decidua), while it was considered as ‘low’ quality when there was a species mixture, one deciduous of the above mentioned and one coniferous (pine, Pinus sylvestris; silver fir, Abies alba).

Exploratory behaviour: To measure this we performed a novel environment test in the indoor facilities of the Netherlands Institute of Ecology (NIOO-KNAW). During the novel environment test birds were tested individually in a sealed test room (4.0 x 2.4 m and 2.3 m high) under artificial light with five artificial wooden trees. Behaviours were observed through an one-way glass window. All movements the focal bird made were noted by an observer. The number of movements in the first two minutes since the bird entered the room was used as measurement for exploratory behaviour (exploratory score). We used the total number of flights and hops within the first 2 minutes corrected for June date (see Dingemanse et al. 2002) as our measure of exploratory behaviour.

Data Processing

All data was analysed and processed in R statistical software using lme4 package. The occupancy data and habitat quality data were binary data (0 or 1). Reproductive measures, disturbance and exploratory behaviour data were continuous data. In order to make laydate continuous data, we calculated the number of days the first egg was laid after 1st of april.