Long-Term Population Trends of House Sparrow and Eurasian Tree Sparrow in Spain

Simple Summary: Urban areas are constantly increasing, as well as human activities related to them. This can lead to an effect in bird populations due to the alterations caused in nature. According to the latest national censuses in Spain, House Sparrow and Eurasian Tree Sparrow populations are decreasing. In this study, we used ringing station data from Spain to assess the tendency of both populations over a two-decade period of time. We also tried to determine the relation of these two populations and their breeding success with the habitat composition. The data analysis conﬁrmed the decreasing trend in both species. However, Eurasian Tree Sparrow showed signs of increasing presence in urban areas, contrary to our predictions. Furthermore, the productivity of both populations remained stable over the study years. Abstract: Urban areas are constantly increasing, which can cause an effect in bird populations since human activities lead to nature alterations. Populations of House Sparrow ( Passer domesticus ) and Eurasian Tree Sparrow ( Passer montanus ) have been decreasing in Spain according to the latest national censuses in Spain. In this study, we tried to assess the population trend over more than two decades using ringing data from Spanish constant effort sites, as well as to determine the population and breeding success proxy in relation to habitat composition at landscape level. We analysed the data and conﬁrmed the decreasing trend in the two species. However, Eurasian Tree Sparrow showed signs of increasing presence in urban areas. Furthermore, the productivity remained stable over sampling sites and years, meaning that the causes of the decreasing populations are affecting both adult and juvenile individuals.


Introduction
The increase in urbanized areas impacts on biodiversity by reducing available natural habitat or fragmenting it [1,2]. Urban environments differ from natural ones in several ways, allowing some species to exploit this new habitat. First, the microclimate that big cities generate is advantageous for many species, usually the generalist ones, due to the heat island effect [3]. This not only alters the phenology of different species but also their ability to survive hard winters, increasing its fitness [4]. Urban areas also provide great amounts of food for some species, up to the point that there is usually a constant supply, due to feeding on organic waste or supplementary human feeding that occurs in urban areas [5,6]. Nevertheless, it usually is low-quality food, which may result in a diet deficiency and even differences in cell composition between rural and urban populations [7,8]. Again, generalist species are the ones who would be able to take advantage from this since they are not so demanding in their ecological niche [9]. In addition, there are low levels of predation since urban areas generally lack natural predators. This implies that there are more weak Birds 2023, 4 160 competitors in urban than in rural areas, where stronger competitors are dominant [10]. Nonetheless, feral and free-ranging cats are highly present in urban and suburban areas, which can cause individual losses or sub-lethal effects such as behavioural changes that can lead to a reproductive decrease or less food intake [11,12]. Pollution is another factor to take into consideration: society is constantly releasing chemicals which damage water, atmosphere, and ground [13]. Moreover, urban areas are more populated, resulting in more pollution than in rural ones [14]. Birds in cities are often in contact with humans. The disturbance created by people to these animals changes considering the species (bigger species display stronger alert behaviour at longer distances than smaller species) but it usually causes a decrease in foraging time, reproduction and parental care [15].
Among all urban bird species that share habitat with humans, sparrows are a representative group due to their large distribution around the planet and vast number of species. They are deeply connected to human structures as well as to their activities [16].
House Sparrow (Passer domesticus) is considered a sedentary species, especially in the Mediterranean area [17]. It is a well-adapted species to urban and suburban areas and it has a great ability to respond to stress in these sites [18]. It is not very selective in terms of habitat and diet. House Sparrow has been associated with human activities such as agriculture since long before ancient times [19]. It adapted to take advantage of artificial habitats such as farmlands and cities and it tends to disappear when humans are not present [20]. Since they depend on human activities, they are usually found in human inhabited areas. This species can be typically found in barns, vegetated patches, and residential areas, especially where gardens are present [21]. Such places provide them with cavities to nest (in buildings), as well as enough food provisioning to feed their nestlings (insects in parks) [22]. The House Sparrow used to be really abundant, but it has decreased markedly in recent decades [16][17][18][19][20][21][22][23]. This phenomenon has been reported not only in the countryside but also in urban environments [24][25][26]. The population of House Sparrow in Spain has had a 18% decrease in the last 25 years according to national censuses, even though the distribution range over the country has remained almost the same [27].
Eurasian Tree Sparrow (Passer montanus; hereafter Tree Sparrow in short), on the contrary, is thought to be related to wetland habitats [28,29]. It is regarded as a more natural species, not so-well adapted to cities as the House Sparrow [30], preferring parks or suburban areas when in cities, distancing from places such as main roads or commercial centres [31]. Tree Sparrow has undergone changes in population in European cities since the 20th century, when it started to colonize urban areas [32]. It has been expanding across Europe, but there is evidence that some countries have had declining trends in the last few years [33]. In Spain, decreasing numbers (−13.52%) have also been reported for this species for the last two decades [34,35]. There is no evidence of negative interactions between the two species. Moreover, House Sparrow tends to disappear from locations where Tree Sparrow cannot be found [36], and some authors point out that Tree Sparrow benefits from the House Sparrow absence [33].
The cause for the descending tendencies in both sparrow species is not clear yet. There may be not only one but several reasons, such as the loss of favourable breeding sites by new and improved construction techniques and materials, or the present management model of urban parks [21,37]. It all may affect to the productivity of the species or to their survival. Productivity is usually defined as the number of offspring a population produces each season. Productivity numbers may change depending on different factors which can affect both adult and juvenile numbers. For instance, the direction a nest faces or its height could determine the rearing success of the chicks [38,39]. In the same way, the amount of available food for nestlings would affect their chances of survival [40]. Therefore, productivity can be used as a proxy of population composition changes.
Constant effort ringing stations (CES) provide a useful approach of ringing data. The conditions must remain stable along the years, during the same breeding periods and temporal scales [41][42][43]. In this way, it is possible to compare population tendencies at different locations undergoing different selective pressures [44,45]. Ringing data could Birds 2023, 4 161 clarify the reason for the decline, since it allows us to identify the age of the captured birds and, thus, we would be able to estimate a productivity index for each species over a long period of time.
The aim of the present study is to determine whether there is a generalized population decrease in the ringing data of both the House Sparrow and Tree Sparrow in Spain that would reflect a global trend not influenced by local conditions, or if population declines are more restricted phenomena caused by regional or local factors such as a decrease in productivity in some geographic areas or by the difference in habitat structure. To attain that goal, we used ringing and productivity data from different constant effort ringing stations (CES). We studied (i) population trends across Spain for 27 different locations over a 25-year period and (ii) the productivity index in those locations in the same time period.
As stated above, in Spain, both species are decreasing their number, so we predict that there will be a decline in the number of captured birds over the 25-year period under study [27,34,35]. We also expect that House Sparrow will be more frequent in urban areas since they are well adapted to cities and human activities, whilst Tree Sparrow, as it is considered a non-urban species, will be associated with more natural areas. In addition, since there seems to be an overall reduction in the population of sparrows, we predict that productivity will be affected, so we will find decreasing productivity values.

Sparrow Data
We recorded data from a network of 27 constant effort ringing stations ( Figure 1, Table 1) during the breeding season with regular trapping intervals following standardized CES protocols (no bird calls, 19 mm mesh size, sampling effort five hours after sunrise) [41][42][43]. The number and locations of the mist nets, as well as their technical characteristics, remained constant across the years and within localities. Accordingly, this sampling protocol allowed us to obtain standardized data that was useful in estimating breeding population trends and allows the identification of changes in juvenile and adult proportions, thus permitting estimates of variation in productivity over years and longterm trend assessment. The data analysed were provided by two different CES programs: PASER (SEO/BirdLife) and EMAN (Aranzadi) [42,43]. Both programs operate during the breeding period (April-July) with 10 sampling days in periods no longer than 10 days apart.    To determine how urbanization affected sparrow populations, we characterized the habitat in a 1 km 2 plot around each ringing station with different parameters: vegetation, buildings, pavement, and water following Liker et al., 2008 [46]. A grid containing 100 cells of 100 m × 100 m was plotted over a satellite picture and a percentage (0-100%) of the Birds 2023, 4 163 previous parameters (vegetation, buildings, pavement, and water) was assigned to each cell. As two different observers analysed different ringing stations, a repeatability test was carried out (r = 0.9993), thus enabling us to merge both scores.

Statistical Methods
To assign an urbanization index to each ringing station, we performed a principal component analysis (PCA). Component 1 of the PCA (PC1) was interpreted as the urbanization index (eigenvalue = 2.7320; explained variance = 0.6830; factor loadings: vegetation = −0.5671, buildings = 0.5821, pavement = 0.5827, water = −0.0057; see Appendix A). Thus, positive scores in PC1 were related to more urbanized areas whilst negative scores were related to more natural areas (Table 1).
A total of 5080 records of House Sparrow and 2553 records of Tree Sparrow were analysed. Birds were marked and dated as juvenile (hatched individuals from that breeding season) or adults (breeding individuals) following Demongin, 2016 [47]. In order to standardize the effort, since not all ringing stations set up the same number of mist nets, an index of capture was calculated by dividing the number of individuals/meters of mist net.
We performed generalized linear mixed models (GLIMMIX, lmer library in R: linear mixed-effects model) to determine differences in the capture index, which was used as the response variable. We used the urbanization index and year as covariates and ringing station as a random factor.
To analyse changes in productivity in all those years, we created a productivity index by subtracting the number of adults to the number of juveniles per year (productivity index: juveniles-adults). A negative index would indicate a higher number of adults, a positive one would indicate more juveniles, and 0 would indicate an equal number of both age groups. We performed another set of GLMM with productivity as a response variable, year and urbanization index as covariates, and ringing station as a random factor.

Population Trends
We found a significant decrease in the capture index for both House Sparrow (p < 0.0001) and Tree Sparrow (p = 0.0037) over the study period (1995-2021) ( Figure 2, Table 2).  However, we found marginally significant differences for House Sparrow between the urbanization index and the captures (p = 0.0683). Capture index increased more in urbanized areas than in natural ones. In the case of Tree Sparrow, we found significant differences in the capture index in relation to urbanization (p = 0.0043) (Figure 3, Table 2).  However, we found marginally significant differences for House Sparrow between the urbanization index and the captures (p = 0.0683). Capture index increased more in urbanized areas than in natural ones. In the case of Tree Sparrow, we found significant differences in the capture index in relation to urbanization (p = 0.0043) (Figure 3, Table 2). However, we found marginally significant differences for House Sparrow between the urbanization index and the captures (p = 0.0683). Capture index increased more in urbanized areas than in natural ones. In the case of Tree Sparrow, we found significant differences in the capture index in relation to urbanization (p = 0.0043) (Figure 3, Table 2).

Productivity
We found no significant differences in productivity between the years in neither House Sparrow (p = 0.1872) nor in Tree Sparrow (p = 0.1479). Similarly, there were no significant differences in productivity depending on the urbanization index with which the ringing station was characterized (for House Sparrow, p = 0.7627; for Tree Sparrow, p = 0.8851) (Figures 4 and 5, Table 2).

Productivity
We found no significant differences in productivity between the years in neither House Sparrow (p = 0.1872) nor in Tree Sparrow (p = 0.1479). Similarly, there were no significant differences in productivity depending on the urbanization index with which the ringing station was characterized (for House Sparrow, p = 0.7627; for Tree Sparrow, p = 0.8851) (Figures 4 and 5, Table 2).

Productivity
We found no significant differences in productivity between the years in neither House Sparrow (p = 0.1872) nor in Tree Sparrow (p = 0.1479). Similarly, there were no significant differences in productivity depending on the urbanization index with which the ringing station was characterized (for House Sparrow, p = 0.7627; for Tree Sparrow, p = 0.8851) (Figures 4 and 5, Table 2).

Discussion
Our long-term study of Spanish House Sparrows and Tree Sparrows strongly suggests a continuous decline in the population of both species over the study years, not only in highly urbanized areas but also in the more natural ones. What we can gather from our results is that the decreasing population trend in both urban and natural areas means that there must be a group of general causes affecting the populations rather than particular local conditions. Although different limiting factors could influence the decrease in the urban population, our data indicate that it is not related to a decrease in productivity in any area, since its index remains the same along the years studied and along the urbanization index.
The decreasing situation of the two species could be due to many different factors since there seems to be no single reason that explains the observed situation entirely; these are discussed below. It is true that cities provide some benefits for species, like a constant food supply; however, there must be a cause for such decline [6]. Some authors have pointed out that the disappearance of favourable habitat and lack of breeding sites may be one of the reasons [21,37]. In relation to the absence of breeding locations, both species lean towards using the space under roof tiles to nest in urban areas [48]. It was shown by Shaw et al. in 2008 that more sparrows can be found in low socioeconomic status locations than in wealthy areas since construction flaws are almost non-existent in affluent regions, making it more difficult for them to nest [37]. To compensate for the lack of nesting spaces, nest-boxes are usually suggested. Interestingly, Tree Sparrows do use nest-boxes but House Sparrows are more indifferent, only using them in some cases where sub-optimal nesting sites had been lost [48,49].
In addition, a descending number of feeding resources such as arthropods in cities may be associated with the decreasing trend of sparrows, since nestlings are fed with them, a lower quantity of insectivore diet could lead to a reduced fitness for the individuals [50][51][52].
Competition with other species could also be a reason leading to the population decrease, due to the use of similar resources and aggressive behaviours to defend them, such as invasive species, e.g., Monk Parakeets (Myiopsitta monachus) [53][54][55]. Predators are another factor to take into consideration, even though cities cannot compare to natural areas when talking about predators, feral cats and free-ranging domestic cats are one of the main causes of biodiversity loss in urban areas [11,12,56].
Another key point is contamination of the environment, such as electromagnetic radiation, additives in fuels, or pollution in the system (atmosphere, ground, and water) [13,57,58]. Probably a combination of these factors and/or others unidentified may be causing the observed decline.
At the same time, Tree Sparrows were thought to inhabit more natural areas, but our research shows how captures increase with more urbanized ringing station locations in contrast to what we had previously predicted [30,33]. House Sparrows are more aggressive than Tree Sparrows, excluding the last ones from favourable nest-sites. Therefore, the decrease in House Sparrow population could lead to an increase in the Tree Sparrow population in urban areas thanks to the absence of the former [48,59]. Nonetheless, Tree Sparrows are influenced by human activity, preferring urban places where it is not so crowded and where they can find a great portion of vegetation. In cities, they prefer places where they can find natural resources, as well as holes in low buildings to nest [31]. The increasing urban population of Tree Sparrows found in our results can reflect occupation on more natural areas within urban cities, for instance, private gardens in residential places [48].
Our data also show that, even with decreasing trends in population, productivity does not change neither along the years nor along the urbanization index. This means that whatever is causing the declining trend in House Sparrow and Tree Sparrow populations, it affects both adults and juveniles. Fitness would be reduced, meaning they would have less offspring. Related to the invariable productivity, birds' breeding seasons are longer in urban areas than in natural ones, having more time to lay more eggs, compensating for their potentially shorter lifespan [60]. Additionally, along these lines, a greater amount of food in urban areas may lead to a higher number of chicks per breeding season, though the quality food is not good enough for a large nestling's lifespan [7,60].
According to Spanish national censuses, and in view of the arguments above, we can conclude that House Sparrow and Tree Sparrow populations are decreasing in Spain. There is no single cause for this event, but it seems that there may be multiple factors involved. Nonetheless, in some North Palearctic areas, there is evidence that Tree Sparrow populations are increasing [33]. In a widely distributed species, such as Tree Sparrow and House Sparrow, population trends in the peripherical areas of the Palearctic distribution do not necessarily have to match with the trends throughout the entire distribution.
Not only that, Tree Sparrows (which used to settle in more natural areas) now unexpectedly seem to be expanding to cities and urban areas. Again, we are not aware of the reason why, but it could be related to the occupancy of empty niches due to the decrease in House Sparrow numbers. However, our data show how productivity remains the same over the study years and locations, which means that the cause of the decline in both populations seems to affect not only adults but also juveniles.
This research shows how constant effort sites are a useful tool to assess trends in bird populations thanks to large periods of time records. Nevertheless, the knowledge of actual reasons for particular declines requires specific approaches and studies that CES data might lack. Further research should be carried out in order to be able to properly identify the causes for the decline of both species' populations.  Institutional Review Board Statement: All the trapping was carried out over the entire study  following Spanish Nature and Conservation policies.
Data Availability Statement: All data are available at the Oficina de Anillamiento from Aranzadi database and the Centro de Migración de Aves, SEO/BirdLife database. Figure A1. Graphical plot of the principal component analysis. "Veg" refers to vegetation, "Pav" refers to pavement, "Bdg" refers to buildings, and "Wat" refers to water.  Figure A1. Graphical plot of the principal component analysis. "Veg" refers to vegetation, "Pav" refers to pavement, "Bdg" refers to buildings, and "Wat" refers to water. Component 2 of the PCA (eigenvalue = 1.1746) was not considered since only an urbanization index was wanted. Water sources, which is what PC2 refers to, was not relevant to this study.