Long-term changes in chick survival rate and brood size in the Grey Partridge Perdix perdix in Poland

ABSTRACT

Data from 4801 Grey Partridge Perdix perdix broods collected in Poland during1987–2013 show that, over the 27 years of study, average brood sizes declined from 10.2 to 6.7 and chick survival fell from 57% to 34%. A probable cause of these trends was a 2.5 fold increase in the use of pesticides over the same time period.

The abundance of Grey Partridges Perdix perdix has declined considerably in western Europe over the last few decades, and one of the most important reasons was a marked decrease in chick survival rate and brood size, due to the intensification of agriculture (Potts 1986, Kuijper et al. 2009, Gée et al. 2018, PECBMS 2019). In Poland, similar demographic trends in this species were observed in the late 1970s and 1980s (Panek 1991). However, it was later found that the survival rate of Grey Partridge chicks in Poland at the end of the 1980s was relatively high (Panek 1992). Furthermore, the chick survival rate only slightly decreased in the 1990s, when the long-term collapse in Grey Partridge populations occurred throughout Poland, mainly due to the deterioration of nesting success and female survival during the breeding period (Panek 2005). Therefore, the Grey Partridge was mentioned as an exemplary species showing dissimilarities in the ecology of farmland birds between western and central-eastern Europe caused by differences in the character and intensity of agriculture (Tryjanowski et al. 2011). It can be expected, however, that the observed progressive increase in the intensity of Polish agriculture (Igras 2014) will result in a further decrease in the survival of partridge chicks and in the size of broods of these birds. The aim of this paper is to test this prediction based on long-term data from the whole country.

Data on the size of Grey Partridge family coveys after the breeding season were collected in the years 1987–2013 as part of a countrywide game monitoring programme conducted by the Research Station of Polish Hunting Association in Czempiń (Panek 2006). In the years 1987–90, fieldwork was carried out only in three areas, but evenly distributed in the country (near Poznań in the west, between Warszawa and Łódź in the centre and near Lublin in the east). During 1991–2013, fieldwork was carried out across further areas, so during the 27 year study period, data were collected in 68 different areas (each typically about 50 km²) and individual areas were actively surveyed for 1–23 years. Notes were kept by trained co-workers, i.e. local hunters and game managers. They registered the size and age composition of all Grey Partridge coveys observed in August while watching crop stubble fields or searching crops with dogs, as well as other opportunistic sightings. Young and adult birds were distinguished on the basis of their size and plumage colouration (Potts 1986). If circumstances indicated that only part of a covey could be seen (e.g. it stayed on the edge of high vegetation), its observation was omitted. In total, data on the size of 4801 Grey Partridge broods were recorded, and annual samples ranged from 1 to 45 in individual areas and from 54 to 403 in the whole country.

The intensity of agriculture in Poland shows a general decreasing trend from west to east (Igras 2014). In order to control for potential regional effects, the country area was divided into three regions: west, centre and east, along the meridians 18° and 21°. Changes in the size of Grey Partridge broods in time and space were analysed using general linear models (GLM, Statistica software) with the annual arithmetic mean brood sizes in the three regions as a dependent variable and two explanatory variables: year (continuous) and region (category). Next, the interaction between year and region was included in the above model to test differences in the course of temporal changes in the brood size between the three regions (i.e. differences between the slopes of regressions describing these relationships). Furthermore, the survival of chicks and the size of broods in the Grey Partridge in Poland were found to be influenced by weather conditions (Panek 1992), so annual fluctuations in the weather could disturb the picture of potential long-term trends in these demography parameters. Therefore, data about mean daily temperatures during the chick rearing period were used in the next step of analysis. Namely, another GLM was conducted for the annual mean brood sizes in the three regions in relation to two continuous explanatory variables, i.e. year and temperature, with region as a category. Partial regression coefficients (r_p) were calculated. Mean daily air temperatures in the main period of small chick occurrence in Poland (the second half of June and the first half of July; Panek 2002) were obtained from 9 meteorological stations located throughout the country. Annual average temperatures in individual regions were calculated based on the data from three meteorological stations located in the north, centre and south of each region. The chick survival rate was computed from geometric mean brood size (GM) according to the formula: 3.665 (0.882GM)^1.293 (Potts 1986, Panek 1992). The results of these calculations may be affected by changes in the size of clutches, however, the mean clutch size and hatching rate in the Grey Partridge in Poland were almost equal in the 1970s and at the turn of the 1990s and 2000s (Panek 2002), so it is likely that no clear trends in these parameters occurred during the period of this study.

The size of Grey Partridge broods in Poland in the years 1987–2013 ranged from 1 to 22, and the annual means for the three regions of the country varied from 4.8 to 11.8 (Figure 1). The annual mean brood size did not significantly differ between the regions (F = 2.360, df = 2, 77, P = 0.1), but showed a negative trend over the study years (r_p = –0.6196, F = 47.97, df = 1, 77, P < 0.001). Moreover, no significant region-year interaction was found (F = 0.6094, df = 2, 75, P = 0.5) indicating the lack of clear differences in the rate of decline in the brood size between the three regions. The mean brood size in the country in the last three study years (6.7, N = 181, sd = 2.4) was 1.5 times lower than in the first three years (10.2, N = 326, sd = 4.0).

Figure 1. The mean brood size in Grey Partridges in the three regions of Poland in the years 1987–2013 (west: r = –0.6170, df = 25, P < 0.001, y = –0.137x + 281.6; centre: r = –0.6706, df = 25, P < 0.001, y = –0.099x + 206.1; east: r = –0.6098, df = 25, P < 0.001, y = –0.099x + 206.3).

The annual mean temperatures in the three regions of Poland during the main period of chick occurrence varied from 14.7 to 21.7°C, while the national averages ranged from 15.1 to 21.1°C and showed a non-significant increasing tendency (r = 0.3208, df = 25, P = 0.1). In the GLM analysis, the annual mean brood size in individual regions decreased with the study years (r_p = –0.6959, F = 71.38, df = 1, 76, P < 0.001) and increased with temperature (r_p = 0.4176, F = 16.06, df = 1, 76, P < 0.001), and moreover the regional effect was slightly significant (F = 3.607, df = 2, 76, P = 0.03). The adjusted mean brood size calculated by the model for the average values of continuous variables was 7.6 in the west, 8.1 in the centre and 8.3 in the east. The calculated survival rate of Grey Partridge chick in Poland in the years 1987–2013 ranged from 29 to 61% (Figure 2), and the average value for the last three study years (34%) was 1.7 times lower than in the first three years (57%).

Figure 2. The survival rate of Grey Partridge chicks in Poland in the years 1987–2013.

During the 27-year period at the turn of the 20th and 21st centuries, a marked decrease in the mean brood size in the Grey Partridge was found in Poland, indicating a clear decline in the chick survival rate in this species. Moreover, the mean brood size was also positively influenced by air temperature. This suggests that the rise in temperature within the context of global climate change (IPCC 2015), which probably occurred also in the chick rearing period in Poland, could result in some limitation of the observed decrease in Grey Partridge chick survival. In fact, when the temperature variable was taken into account in the GLM analysis, the negative partial correlation coefficient describing the temporal trend in the mean brood size was lower than that obtained in the model without the effect of temperature.

The growing use of pesticides is considered to be the main factor causing the decrease in the chick survival rate and the number of offspring in Grey Partridges in Europe (Potts 1986, Kuijper et al. 2009). These substances destroy insects and wild plants which are hosts of many insect species, and have led to a clear reduction in the insect food resources available to partridge chicks (Rands 1985, 1986, Potts 1986, Boatman et al. 2004). In Poland, the sale of pesticides increased almost 2.5 times between the early 1990s and 2013 (GUS 2014 and the earlier volumes of the series). So, this growth in pesticide use could be the main cause of the observed decline in the survival of Grey Partridge chicks in Poland. Moreover, when the impact of temperature on the size of broods was taken into account in the GLM model, the regional effect also turned out to be significant. The lowest average brood size was found in the west of the country, while the highest in the east, which is consistent with the regional differences in the intensity of agriculture in Poland (Igras 2014). However, the regional changes in the mean brood size could be also related to the possible west–east upward trend in the size of partridge clutches (Panek 2002). Another factor affecting the abundance of insect food availability was found to be the structure of the agricultural landscape, i.e. the degree of crop fragmentation and especially the occurrence of uncultivated pieces of land covered with spontaneous permanent vegetation, which are important refuges for some insect species (Panek 1997). During recent decades, a progressive increase in the average area of farms and a modification of the crop structure has been recorded in Poland (GUS 2014 and the earlier volumes of the series). Therefore, it is likely that the agricultural landscape in the country was recently subjected to gradual simplification and this phenomenon could be an additional reason for the observed decline in the survival rate of Grey Partridge chicks.

To limit the indirect negative impact of pesticides on the survival of Grey Partridge chicks and the abundance of this species, a modified method of using these substances was developed, which consisted in omitting cereal field margins during spraying (Rands 1985, Potts 1986). The application of this method resulted in a clear increase in the abundance of insects preferred by partridge chicks, and consequently also in their survival rate and brood sizes (Rands 1985, 1986, Potts 1986, Chiverton 1999). Furthermore, the unsprayed field margins (named ‘conservation headlands’) in combination with other habitat management projects and predator control actions typically allowed for a clear increase in the number of Grey Partridges (Ewald et al. 2010, Aebischer & Ewald 2012, Sotherton et al. 2014). The decrease in the survival rate of Grey Partridge chicks found in this study indicates that the use of pesticide spraying modifications should be beneficial for these birds also in Poland. The quality of farmland habitat could be improved within agri-environmental schemes, but recently such treatments have not been widely used in Poland, and the most popular variants (e.g. the maintenance of traditional grassland, the protection of soils and water bodies) were not targeted on animals living on agricultural land (Pawlewicz & Bórawski 2013).

The indirect effect of pesticides on food resources and breeding performance was demonstrated or at least suggested also in some other farmland birds (Campbell et al. 1997, Boatman et al. 2004, Bright et al. 2008). It seems that the Grey Partridge, and specifically the size of broods in this species, can serve as an indicator of food resource quality for insect-eating birds living in the agricultural landscape.

Acknowledgements

I am grateful to all hunters and game managers for their help in data collection, and to my colleagues from the Research Station PHA in Czempiń for their valuable comments to the manuscript.

ORCID

Marek Panek http://orcid.org/0000-0002-0915-1636