The influence of climate change on brood size and breeding time in three tit species in Sweden, 1962–2019

climate change is expected to affect many biological systems, including the timing of seasonal events such as breeding in birds. In this study, I investigated how brood size and timing of reproduction (measured as median date for ringing of broods) changed for three tit species in Sweden 1962–2019 using data reported to the Swedish Bird Ringing Centre. The brood size for the Coal Tit Periparus ater increased from 7.74 between 1962–2001 to 7.98 young between 2002–2019, while no change was detected for Crested Tit Lophophanes cristatus or Willow Tit Poecile montanus . The largest biological effects were seen for timing of reproduction, with Coal Tits and the Crested Tits initiating breeding about seven days earlier at the end of the study period compared to the 1960s. The Willow Tit data also suggested earlier breeding, but only by about two days across the study period.


Introduction
The Intergovernmental Panel on Climate Change (IPCC) predicts increasing global warming during the next 100 years (IPCC 2021).The rate of increase is higher in the Northern hemisphere, dependent on the larger land mass and smaller ice cover (IPCC 2021).An increased average temperature induces many changes in biological systems and have profound effects on both plants and animals over large areas.Important changes in Europe include different timing of seasonal events, such as an advancement of spring and changed phenological timing of adult insects and larvae (Visser et al. 1998).Birds relying on such food sources need to adjust to changed timing of their prey or risk a phenological mismatch, leading to reduced breeding output (Miller-Rushing et al. 2010, Saalfeld et al. 2019).
In an earlier study using ringing data on four tit species (Paridae) in Sweden, Ryttman & Karlsson (2010) found that Coal Tits Periparus ater, Crested Tits Lophophanes cristatus, and Marsh Tits Poecile palustris had advanced their breeding over a 40-year period (1962-2001), while Willow Tits Poecile montanus had not.As another two decades have passed since that study, I wanted to return to this subject and investigate whether the pattern with earlier breeding continues and-if so-how rapid this change is for the Coal Tit, Crested Tit, and Willow Tit, breeding in the same habitat.

Material and methods
The underlying data for both the previous and this study are ringing data of broods reported to the Bird Ringing Centre at the Swedish Museum of Natural History in Stockholm.The centre oversees all ringing conducted in Sweden, and brood data can be separated from other types of bird ringing data with the use of specific codes in the database.In total there were 4,967 broods available, of which 3,884 were from Coal Tits, 644 from Willow Tit, and 455 from Crested Tit across the years 1969-2019 (Table 1).
Reproductive timing was measured by estimating the median date for when broods were ringed for each year from 1962 to 2019, except for Crested Tit, where no brood was ringed in 2019.When no specific date could be determined I used a mean between the two dates for the median, or the most likely date (the date when most broods were ringed).
In addition, the mean number of ringed young was estimated for the three species and a comparison between the two periods (1962-2001 vs 2002-2019) was made.

BROOD SIZE
The number of young per brood increased for Coal Tit, from a mean of 7.74 during the years 1962-2001 to 7.98 in the period 2002-2019 (t = 2.92, n = 3,882, p = 0.003).

TIME OF BREEDING
All species advanced their breeding, measured as the time for ringing young-which is done at nearly the same development stadium of the young, during the study period.Regression analyses of median date over year all resulted in negative slopes (b-values), corresponding to an earlier breeding time by six to seven days today compared to the early 1960s for Coal Tit (r = 0.418, t = 28.6,n = 58, p < 0.001; Figure 2a) and Crested Tit (r = 0.285, t = 6.33, n = 57 p < 0.001; Figure 2b).The Willow Tit advanced its breeding by only about two days, but this estimate is also significant (r = 0.117, t = 2.99, n = 58, p = 0.003; Figure 2c).
The mean temperature during 1962-2019 in April in Stockholm increased by over 3 °C since the 1960s (Figure 3).

Discussion
The data presented here show a clear pattern in all three tit species that timing of breeding has advanced significantly over the 57-58 years of the study.Furthermore, the rates of advancement are fairly constant across the time series, albeit with some interannual variation.It is important to note that brood ringing occurs at specific development stage of the nestlings and is therefore not a variable that has varied over time.Thus, a likely explanation for a progressively earlier ringing of broods is climate change-driven warmer temperatures in Sweden, resulting in an increasing advancement of spring and corresponding selection for earlier breeding among the tits.
In a long-term nest-box study on Coal Tits from Örebro in south central Sweden in 1949-2012, Schölin & Källander (2013) found that there were no trends in laying date in the first 27 years, but an increasing advancement in the second 28-year period; this advancement correlated with increasing temperature in April.The same study also found a similar correlation between advanced laying date and increased temperature over time in Blue Tits Cyanistes caeruleus, where mean temperature in April increased by 3 °C in their study region (Schölin & Källander 2012).
An interesting result is the increasing number of young per brood observed for the Coal Tit (Table 1; Ryttman & Hall-Karlsson 2010).Schölin & Källander (2013) reported an average of 7.78 ringed young Coal Tits per brood from a mean of 8.55 eggs per clutch (n = 105) during 1949-2012, which is comparable to clutch sizes reported here.The reason for the increased brood size of Coal Tits is difficult to explain.The two other tit species do not show the same pattern, despite breeding in similar habitats and showing advanced breeding times.Potential artefacts could be differences resulting from sampling different geographic regions of Sweden, which was not investigated here, or that nest-box placement affects breeding output.I have previously shown increased brood sizes in another nest-boxing species, the Eurasian Wryneck Jynx torquilla between 1962 and 2012 (Ryttman 2003(Ryttman , 2014)), where I considered nest-boxes being placed in more suitable areas in the 1990s, when many new  1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 2010 2014 2018 Median date (1=1 May) Mediandatum (1=1 maj) Year År ringers started projects for studying the breeding of Eurasian Wrynecks.The Swedish Bird Monitoring (www.fageltaxering.lu.se) shows different results in the different types of areas where monitoring is carried out.In point counts, which represent survey sites that are selected freely by the observers, all tits have decreased (p < 0.01; Green et al. 2019).Across the standard routes, which represent surveys in predetermined areas representative of the overall habitat composition, Green et al. (2019) show that the Coal Tit has a stable population while the Crested Tit increases (p < 0.001) and the Willow Tit decreases (p < 0.05).The Coal Tit and Crested Tit have larger populations in southern Sweden while the Willow Tit has a distribution which is more northerly.The less pronounced reaction to the changing climate in Willow Tit could potentially affect its decreasing population, perhaps depending on the disadvantage of missing the most suitable time for feeding their young (cf. Visser et al. 1998).However, such a disadvantage of mistiming should be detectable as fewer chicks alive at the time of ringing, and this was not the case.Fewer decomposing birch Betula spp.trees in forests may pose a constraint to finding suitable nesting sites, but that should equally affect the Crested Tit.
Ornis Svecica (ISSN 2003(ISSN -2633) is an open access, peer-reviewed scientific journal published in English and Swedish by BirdLife Sweden.It covers all aspects of ornithology, and welcomes contributions from scientists as well as non-professional ornithologists.Accepted articles are published at no charge to the authors.Read papers or make a submission at os.birdlife.se.