1. Introduction
The reproductive life history represents the overall trade-off between investment in egg and clutch and timing of reproduction of important events in the reproductive process of biological organisms, with a particular focus on the mechanism of why different species in the same environment or the same species in different environments may develop different reproduction strategies [
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11]. Reproduction, the most concerned content of ornithologists, is the fundamental process for life history, and reproductive strategy is the decisive factor that affects the fitness of animals, directly through individual survival and reproduction, population dynamics and continuity [
3,
10,
12,
13,
14,
15,
16]. Large numbers of studies have described the diversity of reproductive life history strategies of birds due to variation of environmental conditions and individual morphology [
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26,
27,
28,
29]. Additionally, the body size and reproductive characteristics in populations of the same species is also influenced by environmental and phylogenetic factors [
30,
31,
32]. Reproductive life history characteristics of birds, such as egg-laying time, egg size, clutch size and clutch volume, are usually considered to be the main indicators to measure reproductive output [
10,
27,
28,
29,
30,
31,
32,
33]. These indicators can effectively reveal the different reproductive traits of birds in different environments due to adaptive trade-off and constraint of physiological mechanisms, which can be interpreted as different reproductive life history strategies [
27,
28,
29,
30,
31,
32,
33,
34]. For example, egg size and clutch size of birds can be adjusted according to the changes of egg-laying date, ambient temperature, food resources and female physical condition [
4,
28,
35,
36].
Life history theory seeks to explain how natural selection and other evolutionary forces shape organisms to optimize their survival and reproduction in the face of ecological challenges posed by the environment [
1,
2]. The trade-off between egg size and clutch size is one of the core principles of life history evolution, in which numerous studies have reported a rather positive covariance between egg size and clutch size: good-quality females may lay more and larger eggs compared to low-quality females [
1,
36,
37,
38]; adequate food resources usually lead to early laying of eggs and larger clutch sizes, instead of larger eggs [
6,
39,
40]. Theoretically, the reproductive output of parent birds that lay eggs at a fixed number can be increased by increasing the size of a single egg, but the physiological constraint hypothesis predicts that the reproductive output will be constrained by the maternal condition, which is specifically manifested in that larger individuals lay larger eggs, while smaller individuals lay smaller eggs due to insufficient energy storage in their bodies [
41]. The body size can partly represent the female's energy storage, so as the body mass increases, the female can supply more energy (egg mass) or lay more eggs [
42]. The size of egg has costs and benefits, and individuals might balance these costs and benefits when allocating resources to the size of an egg [
43]. According to life history theory, females should have some flexibility in the allocation of resources on eggs, including laying the optimal egg size and small clutch size in the case of abundant resources, and laying relatively small egg size but large clutch size in the case of limited resources [
2,
29,
44,
45]. However, the constant number of offspring has been found in some groups. Birds provide care to their offspring during laying, incubating and brooding. As the incubation capacity hypothesis predicts that the maximum clutch size that females can lay is constrained by the incubating capacity (e.g. incubating spots size), this may explain why most species of shorebirds have a relatively invariant clutch size [
46,
47]. Thus, most plovers vary their reproductive output by varying their egg size rather than clutch size [
6].
On the other hand, the morphological constraint hypothesis holds that the egg length or width was constrained by the size of female partial morphological features. When the egg width is constrained, females can increase their energy investment by laying longer egg, for example, lizards can lay longer eggs to increase their energy supply to eggs as the tail base width of female lizards constrains the egg width [
48]. In the light of this phenomenon, some scholars have put forward the concept of egg shape allometry which states that with the increase of egg size, the growth rate of egg length and egg width is different, and the pattern of allometry in egg shape can be checked by comparing the regression slope between egg length and egg width [
49] Therefore, the size of a single egg can be increased indefinitely. The upper limit hypothesis of maternal constraint predicts that the egg width is physically constrained by the cloaca of bird, whilst the increase of egg size is mainly achieved by increasing the egg length, which is manifested as the egg length allometry [
45,
49].
The Kentish plover
Charadrius alexandrinus is a small shorebird (with a body length of 14~17.5cm) with an extremely large range and multiple geographical races which live in both temperate and subtropical climate zones in four continents. Kentish plovers often inhabit the open and flat coastal beaches, the bare land around saltwater lakes and inland lakes [
50,
51]. The breeding period of Kentish plover is usually from March to August every year with certain variations in different regions. They usually lay 3 oval eggs per nest, accidentally 2 to 4 eggs, with length of 29~35 mm and width of 22~25 cm, mass of 8~11 g [
9]. The female bird lays an egg with 2 day interval until the incubation starts after the last egg is laid. The incubation period is approximately 23 to 29 days [
9,
50]. In China, Kentish plovers breed in coastal areas, inland lakes and near reservoirs in the north, and inhabit all Provinces during migration, and can be seen throughout most of the southeastern coast during non-breeding periods [
54,
55]. Because of their wide distribution, the breeding strategies of Kentish plovers are diverse and have strong plasticity, which attracts the attention of scholars in animal ecology, behavior and evolution [
46]. Variations in breeding strategies are common phenomena in birds living in different distribution ranges [
10,
46,
56,
57]. For example, reproductive characteristics such as egg length and width, clutch size, laying date, incubation rate, growth rate, brood desertion and adult survival often differ distinctively between northern and southern bird species [
58,
59,
60]. Several studies have investigated how reproductive traits vary within a single species as elevation increases and climate becomes more severe [
21,
61,
62,
63]. We hypothesize that, as far as the avian life history variation along altitudinal gradients is concerned, breeding in high-elevation habitat results in shift to shorter life-history strategy within a single species.
At present, the reproductive biology of Kentish plover has been studied mainly in coastal areas, and most of the data come from Europe and North Africa [
50,
64,
65,
66]. In China, the study of Kentish plover reproductive biology mainly focuses on the populations in Bohai Bay [
67], Qinghai Lake [
68]. The populations in inland arid areas of Western China have drawn less attention [
69]. Due to the influence of geography, temperature and environment, organism shows different reproductive strategies among different populations [
45,
47,
68,
70]. In view of the harsh natural environment in inland arid areas of Xinjiang, we hypothesis that Kentish plover may show flexibility in breeding strategies to cope with the changeable environment. Here, we focus on the reproductive traits of Kentish plover for revealing the reproductive strategy of this bird adapting to different localities in arid desert areas of Xinjiang (
Figure 1), where the precipitations vary greatly, but the temperature differences are not significant (
Figure 2). During the early stage of breeding season the climate of Taitema Lake is extremely dry with frequent dust storms. Hence, we hypothesis that, the plovers of Taitema Lake adapt to the poor climate conditions by laying larger eggs and shortening incubation period. According to the physiological constraint hypothesis, we hypothesis that female traits such as body mass or tarsometatarsus length determine reproductive output which is manifested in different egg size and clutch size in different populations. According to the upper limit hypothesis, we hypothesis that larger egg show a more positive allometry in egg shape. Therefore, the primary objective of this study was to provide baseline information about the breeding strategies of the Kentish plover populations in inner arid regions of Xinjiang, one of the important breeding areas for this species in China.
2. Materials and Methods
2.1. Study area
The current study was conducted, during the breeding season of Kentish plover, from April to July of 2019 and 2020. We collected and compared a variety of reproductive traits among three geographically distinct Kentish plover populations of Xinjiang, including Aibi Lake (AL) population in the Northwestern Xinjiang, Taitema Lake (TL) population in the Southern Xinjiang and artificial reservoir group (AR) population around Urumqi City in Northern Xinjiang (
Figure 1).
The Aibi Lake Basin is a closed basin located in the inland area of the Junggar basin in Xinjiang in northwest China (43°380-45°520 N, 79°530-85°020E, with average altitude of 200 m) [
71]. It has a total area of 50621 km
2 where plains make up 25762 km
2. The Aibi Lake, one of the important breeding habitats and temporary rest stations for migratory birds in western China, is under the jurisdiction of the Xinjiang Aibi Lake Wetland National Nature Reserve. The Lake, characterized by its dry and hot weather, high temperature low precipitation and little huaman disturbance, is the largest salt lake in western China, with a water area of 520 km
2. The average annual temperature is 8.3 °C, and the average annual precipitation is 90.9 mm. The average annual precipitation on the surface of the lake is about 95 mm, with an the annual evaporation can reach as high as 1315 mm. Strong winds with a maximum wind speed of 55 m/s are more likely occur from April to June Kentish plover population there usually nesting on saline-alkali land, sand land and gravel substrates in the region [
72].
The reservoir group including Liuchengzi Reservoir and Wushihua Reservoir around Urumqi City is located at the north edge of Boghda Mountain. They are typical plain reservoirs mainly used for aquaculture and irrigation. The specific study sites are located on the southeast bank of Liuchengzi Reservoir (N44.255835°, E87.885218°; with average altitude of 480 m) and the east of Wushihua Reservoir (N44.196667°, E87.741022°; with altitude of about 475 m). Since the distance between the two sites is only 13 km, and there is no difference in female morphology and egg characteristics between the populations form these two sites, we combined those two populations as an artificial reservoir population. This area is close to villages and towns, mostly surrounded by reclaimed farmland or industrial parks with strong human disturbance, such as livestock and road construction which accidentally cause the failure of nests.
Taitema Lake (TL) with a water surface of 300 km
2, is located in the southeast margin of Taklimakan Desert, about 50 km north of Ruoqiang County, Bayingolin Mongol Autonomous Prefecture. It is the terminal lake of three river systems: the Tarim River, the Cherchen River, and the rivers on the northern slope of the Altun Mountains [
68,
73]. The Taitema Lake region has an extreme continental climate, making it extremely arid. The average annual precipitation is 17.4–42 mm, the average annual evaporation is 2500–3000 mm, and the extreme maximum temperature is 43.6℃. Furthermore, above-ground vegetation is sparse. The study area is located on the southeastern shore of the lake (N39.412433°, E88.517855°; with average altitude of 800 m). The breeding habitat of plovers is basically harsh desert composed of desert grassland, sand land and sand dunes which is well-preserved with little human disturbance [
73,
74].
2.2. Data Source
In this study, the regional meteorological data (
Figure 2) came from the National Meteorological Information Center (https://
www.nmic.gov.cn/ en/) of the China Meteorological Administration, with data on temperature and precipitation obtained from the weather station, which is closest to the study region. We calculated the average value every ten days, including daily average temperature (AT), daily maximum temperature (DTmax), daily minimum temperature (DTmin), daily temperature difference (DD) and sum of rainfall (RF) during the breeding season from March 27 to July 24, each year in 2019 and 2020.
2.3. Data collection and traits measurement
We find nests by watching birds with a telescope, binoculars and by searching on foot inside the nesting grounds during the breeding season [
75]. We monitored nests every 3–4 days in each year during the breeding season, until the hatching of eggs or the failure of the breeding attempt. The morphological traits of female parent and its eggs size characteristics (length and width) were measured with a digital caliper (0.01mm) and the weighed with a digital scale (0.01 g) [
75]. GPS data of the nests were recorded. We captured breeding female parent with a walk-in funnel trap placed over the nests that had been incubated for at least seven days [
68]. Female body mass (FBM, g) is considered to be the female's weight after laying all eggs. Female tarsometatarsus length (FTL, mm) is the tarsometatarsus length on the right side of the female. Egg mass (OEM, single egg weight, g), clutch size (CS, the total number of eggs laid in a single breeding period), egg length (EL, linear length at the longest end of the egg, mm), egg width (EW, linear length at the maximum width of the egg, mm), egg shape (ES, ratio of egg width and egg length) were measured for analyzing allometry in egg shape.
We also calculated egg volume (EV, cm
3):
EV =Kv×L×W2, where
Kv=0.5236-(0.5236×2×(L/W)/100),
L=egg length,
W=egg width, and clutch volume (CV, the sum of the volumes of all eggs in a single breeding period, cm
3) [
76]. We estimated laying date (LD) through observation and floating the eggs in luke-warm water [
75]. We used the Julian day, which we calculated as the number of days between the 1st April (i.e. 1) of each year, as the egg-laying date. The time interval between the beginning and the end of hatching of the plover was taken as incubation period (IP). The monitoring of AL population is too difficult and the number of nests was too small, so the data of incubation period could not be obtained by continuous monitoring.
2.4. Statistical analysis
The collected data is sorted out with Excel data. Non-normally distributed data was log10 converted to meet the assumption, i.e. the Kolmogorov-Smirnov normality test and Levene’s variance homogeneity test. We used the One-way ANOVA to make multiple comparisons on the female body mass, tarsometatarsus length, egg size, clutch volume and other traits of the three populations with post hoc Tukey’s tests. For the data that do not meet the above assumptions, we use the nonparametric Kruskal-Wallis test to examine data differences between the three populations with post hoc Steel-Dwass tests (multiple comparisons). We used analysis of independent t-test to examine the annual differences in reproductive traits of each population, the results showed no significant annual differences on reproductive traits among them, so two years of reproductive traits data from each population were combined for subsequent analysis. Model II regression in “lmodel2” package was used to make regression analysis between egg size, clutch size, clutch volume and laying time one by one, as well between egg size, clutch size, clutch volume and the body mass and tarsometatarsus length of their female parent one by one, so as to verify the effects of laying time and female morphology on egg size and clutch size.
For the egg shape allometry, we first make the regression between egg length and egg width. If the slope is greater than 1, there is exists egg shape allometry. Then, residual on the egg length, egg width with clutch size were extracted respectively to eliminate the effects of the clutch size on the egg length and width [
45], and regression was made between the residual egg length and the residual egg width to test whether there is allometry for the egg shape after removing the effects of the clutch size.
Model Selection Analysis was conducted to test the effect of environmental factors and female morphological traits on egg volume and clutch volume. Daily average temperature, daily temperature difference (DD), daily maximum temperature, daily rainfall, laying date, clutch size, the female body mass and female tarsometatarsus length were taken as independent variables to carry out the Multinomial Logistic Model analysis. Populations were included as a fixed index, and the parameter years was entered as a random effect in order to control the interpopulation phylogenetic relatedness.
Akaike information Criterion (AIC) was used to compare models and determine two principles of the best model screening: (1) the minimum AIC value is required; (2) the Dealt value of the model < 2. All data are listed in the form of mean ± standard error (Mean ± SE), and all operations are completed in R.v.3.6.3 software, using the packages ‘‘ggplot’’ and ‘‘gplots’’. Differences were considered significant when P< 0.05.