Changes in the landscape patterns of Black‐necked Crane habitat and its correlation with their individual population numbers during the past 40 years in China

Abstract The landscape pattern of the Black‐necked Crane (Grus nigricollis) habitat in China changed at different spatial scales and long‐term periods due to natural factors and human activities, and habitat reduction and fragmentation threatened the survival of Black‐necked Cranes. The factors driving the habitat landscape pattern and individual population changes of Black‐necked Cranes remain to be studied. In this paper, based on remote sensing data of land use from 1980 to 2020, the changes in landscape pattern and fragmentation of the Black‐necked Crane habitat in China over 40 years were analyzed from two different spatial scales using the land cover transfer matrix and landscape index. The correlation between landscape and Black‐necked Crane individual population was analyzed. The most obvious observations were as follows: (1) Although transformation between landscapes occurred to varying degrees, the area of wetlands and arable land in the breeding and the wintering areas (net) increased significantly from 1980 to 2020. (2) Habitat fragmentation existed in the breeding and the wintering area and was more obvious in the wintering area. (3) The number of individuals of Black‐necked Cranes increased period by period, and habitat fragmentation did not inhibit their population growth. (4) The number of individuals of Black‐necked Crane was closely related to the wetland and arable land. The increasing area of wetlands and arable and the increasing landscape shape complexity all contributed to the growth of the individual population. The results also suggested that the number of individuals of Black‐necked Crane was not threatened by the expanding arable land in China, and they might benefit from arable landscapes. The conservation of Black‐necked Cranes should focus on the relationship between individual Black‐necked Cranes and arable landscapes, and the conservation of other waterbirds should also focus on the relationship between individual waterbirds and other landscapes.


| INTRODUC TI ON
Globally cranes generally have geographically separate breeding and wintering habitats, making it difficult to make protected areas that include both. Furthermore, many crane species also feed on anthropogenically modified land-use types, mainly croplands after harvest Mi et al., 2018). It is important to use remote sensing techniques to show how available habitat types have changed over time. In recent decades, the land-use structure and landscape pattern of the plateau wetlands in China changed due to natural factors and human economic activities Zhao et al., 2020), leading to the fragmentation of the habitats of the Black-necked Crane (Grus nigricollis) and the reduction in suitable habitat areas (Li et al., 2022;Ru et al., 2019). In addition to limiting the movement, dispersal, and exchange of species, habitat fragmentation accelerates the extinction rate of endangered species and increases the loss of species diversity (Haddad et al., 2015;Newbold et al., 2015;Wilson et al., 2016).
Past literature has provided numerous assumptions regarding the relationships between crane habitat-anthropogenic, and studies by several authors have demonstrated that human activities are detrimental to wild species (Bishop, 1996;Li & Li, 2005). However, numerous studies by scholars in Africa and South Asia have shown that agriculture is nevertheless of high value to cranes (Benn et al., 1995;Sundar, 2003). For example, most of the prey of the Black-necked Stork in India is obtained from agricultural fields, and the Blackheaded Ibis also uses agricultural landscapes and densely populated urban landscapes, and Sarus Cranes in South Asia also live mainly in unprotected agricultural landscapes (Koli et al., 2019;Sundar, 2009Sundar, , 2011aSundar, , 2011b. Within regions of high regional population density and cultivation, food diversity and density in arable land make foraging by cranes and storks closely associated with agriculture and can even affect their reproductive success. Moreover, it was also shown in China that Black-necked Cranes use agricultural land and avoid forests in some areas Li, 2014;Tscharntke et al., 2005;Wang et al., 2019;Wu et al., 2020).
Many studies have been conducted on the link between habitat quality and landscape change for the Black-necked Crane, and landscape pattern analysis methods were widely used because of their ability to reflect the habitat's environmental quality effectively (Liu, Ning, et al., 2018;Liu, Wilson, et al., 2018;Chu et al., 2018), the dynamics of habitat suitability (Abdolalizadeh et al., 2019), and the population response to habitat fragmentation (Steffens & Lehman, 2018). The choice of methods for habitat fragmentation analysis mostly favored quantitative analysis using multiple indices characterizing fragmentation and thus exploring the effects on birds (Guan et al., 2022;Steffens & Lehman, 2018), but the following aspects remain to be further investigated. First, analyses of long-term habitat change processes are lacking. Second, there need to be more studies at large scales and research on the factors driving the Blacknecked Crane population at different scales. Third, there is a lack of validation of whether arable land expansion activities in China are detrimental to the survival of Black-necked Cranes.
To address these issues, in this paper, we analyzed the longterm habitat change and habitat fragmentation of the breeding and wintering areas of Black-necked Cranes in China from 1980 to 2020 by using the land cover transfer matrix, landscape level, and class-level landscape index. In addition, the number of individuals of Black-necked Cranes was estimated, and the correlation between landscape pattern changes and the number of individuals of Blacknecked Cranes was analyzed. The analysis was conducted at the maximum spatial scale and with one of the typical wintering areas of Black-necked Cranes (Caohai National Nature Reserve, Guizhou Province) for a more advanced study to show the factors driving changes in the number of individual Black-necked Cranes at different study scales. We hypothesize and verify that agricultural landscapes in China are closely related to Black-necked Cranes and that human activities are not always detrimental to the growth of Blacknecked Crane populations, and we will focus on changes in arable landscapes in addition to wetland landscapes.

| S TUDY ARE A
The breeding areas of Black-necked Cranes are located in Qinghai, Tibet, and the provinces of Sichuan that border Gansu and Qinghai (between 31° N to 39° N and 79° E to 104° E), and the wintering areas are located in the Yarlung Tsangpo River valley in southcentral Tibet, western and northeastern Yunnan, and western Guizhou (between 25° to 30° N and 87° to 105° E) in China (Hou et al., 2021;Li, 1999;Wang et al., 2020). The wintering sites and the breeding sites of the Black-necked Cranes that were recorded during the study were included in the study area, which included 22 urban areas, as shown in Figure 1a (Kong et al., 2018;Li & Li, 2005;Li & Yang, 2003;Zheng & Wang, 1998).
In addition to the study of Black-necked Crane habitats at large spatial scales, the Caohai National Nature Reserve (CNNR) in Weining County, Guizhou Province (between 104°10′ E to 104°20′ E and 26°47′N to 26°52′N) was selected as a typical study area since it has been a classic Black-necked Crane wintering site (Figure 1b, c).
The Caohai wetlands are one of the few natural reserves of subtropical highland wetland ecosystems in China. The natural landscape of the Caohai is severely damaged by intense human activities, resulting in a significant impact on the quality of Black-necked Crane habitats, which caused fluctuations in their wintering population (Li, 2014;Ran et al., 2017;Tao et al., 1997). Therefore, understanding the changes  in the landscape pattern of the Caohai can provide more targeted analyses of the effects of changes in the landscape pattern of the wintering habitat of the Black-necked Crane on their population size.

| Data source and pre-processing
The remote sensing data on land use in this paper were obtained from the Resource and Environmental Science Data Centre of the Chinese Academy (http://www.resdc.cn) (Gao et al., 2019;Liu et al., 2014Liu et al., , 2020Liu, Ning, et al., 2018;Liu, Wilson, et al., 2018) at two different resolutions (1 km, 30 m, respectively), with 6 years of data for each resolution (for the years 1980, 1990, 2000, 2010, 2013, and 2020). The interpreted images used for remote sensing data on land use were mainly Landsat TM, and the accuracy of the interpretation was at least 85%. More data information can be found in Table S1. Except for the seven-year interval from 2013 to 2020, the landscape pattern changes were analyzed at 10-year intervals. The 1 km land-use data were used to analyze the landscape pattern changes between the breeding and the wintering area at the large spatial scale and to compare the degree of habitat fragmentation in breeding areas with that in wintering areas.
The 30 m land-use data were used in the Caohai National Nature Reserve to analyze the long-term habitat change and its impact on the population of Black-necked Cranes at a small landscape scale in a more detailed way. The results of the index analysis at different landscape scales were not compared.
Black-necked Cranes select inaccessible areas for breeding and wintering, such as swamps, lakeside meadows, reed swamps, and river valley swamps, which are far from human disturbance (Kong et al., 2011;Qian et al., 2009;Song et al., 2014). To investigate the effects of changing habitat landscape patterns on the Black-necked Crane, the large-scale regional classification was divided into six basic landscape types based on the ecological habits of the Black-

| Analyses
The transfer matrix reflects the transformation process of the actual state of land cover types from the beginning to the end in the study area. The transfer matrix can be used to analyze the transfer direction and the number of transfers of different land-use types.
The generalized form of the land using a transfer matrix is shown in Equation 1: where S represents the area, n represents the number of land-use types, and i and j (i = 1,2,3…n, j = 1,2,3…n) represent the land-use types before and after the transfer, respectively. S ij represents the area of land-use type i before transfer into land-use type j.
Landscape index analysis is the most commonly used static quantitative analytical method to study the composition and char- AREA_MN, LSI, AI, and COHESION can be applied at the landscape and class levels, while SHDI can only be applied at the landscape level (Leng et al., 2022;Wang et al., 2014). Although there is a positive correlation between AI and AREA_MN and a strong positive correlation between AI and COHESION among these five indices , all three indices were reserved for use in order to quantify habitat fragmentation characteristics and trends from three perspectives: area, connectivity, and aggregation. The landscape analytical software Fragstats 4.2 (McGarigal & Marks, 1995) was used to calculate the indices. The meanings of the indices are shown in Table 1.

Although this study counted the number of individuals in each
Black-necked Crane distribution area in China since 1980, these records of Black-necked Crane population distribution concentrated mainly in the Yunnan-Guizhou plateau, the records of the Black-necked Crane population in Tibet were relatively the rarest Lu et al., 2017;Wu et al., 2020). Therefore, counting the number of individual Black-necked Cranes in the wintering area was relatively easier. Due to the problem of missing and discontinuous data in some areas during the counting process, in this study, the total number of

AREA_MN (Mean Patch Area)
The average area of all patches or a particular type of patch in the landscape (Value range: AREA_MN ≥ 0). The smaller the average patch area, the more fragmented the patch.
Square kilometers LSI (Landscape Shape Index) This index reflects the shape complexity of the landscape (Value range: LSI ≥ 1). The closer the LSI is to 1, the simpler the overall shape. The larger the LSI, the more complex the shape.  were concentrated in the east and north, and landscape changes in wintering areas were partially concentrated in the north (Figure 3).
The transitions of landscape area in the breeding and wintering areas were further summarized by the land cover transfer matrix (Table 3), and it can be seen that in the breeding area, other land types decreased the most. Built up area and arable land increased over time.
Between 2013 and 2020, the area of open water area increased much more than other landscape types (by 3444 km 2 ), and the source of the increase was mainly other land types (by 3090 km 2 ), while the area of beaches and swamps also decreased extensively. In the wintering area, the significant reduction in arable land area in different periods was mainly transformed into other land types (non-wetland habitat types), and the increase in open water area was mainly from the transformation of arable land and other land types (non-wetland habitat types).

| Landscape index analyses for the breeding and the wintering areas
At the landscape level, the Black-necked Crane habitat in the wintering area was more fragmented than in the breeding area, and the Black-necked Crane habitat in the breeding area was better connected and more aggregated. However, after 2010, there was a trend of more obvious fragmentation of habitats in the breeding areas ( Figure S1). At the class level, the shape of all patches in breeding and wintering areas tended to be more complex. However, the aggregation and connectivity of patches in breeding areas were generally better than those in wintering areas. Among all types of habitats, the connectivity and aggregation of open water areas in breeding and wintering areas were good, but there existed a trend of fragmentation, especially in the wintering areas. In addition, the connectivity and aggregation of arable land in the breeding area increased yearly.
However, the connectivity and aggregation of arable land in the wintering area decreased ( Figure S2). The index calculation results and more detailed analysis can be found in the Appendix S1.

| Landscape change analyses in the Caohai National Nature Reserve
The landscape of the Caohai region includes the lake, paddy field, dryland, woodland, grassland, and built up area ( from 1980 to 2020 showed the expansion of the built up area to the southeast in the northeast (Figure 4).
The land cover transfer matrix showed that the grassland and dryland landscapes changed the most in the CNNR (Table S7,   Table S8). The landscape area transitions of the CNNR summarized by the land transfer matrix showed that the lake area increased the most from 1980 to 2010, followed by the dryland, while the grassland area decreased the most. The transformation between landscapes decreased from 2013 to 2020, the arable land area decreased the most, while the building area increased the most during this period (Table 5). More detailed results of the landscape transfer can be found in the appendix files (Table S7, Table S8).

| Landscape index analyses for the Caohai National Nature Reserve
At the landscape level, the landscape connectivity and aggregation in the CNNR have been in good condition since 1980 with a balanced landscape distribution. However, the landscape shape was F I G U R E 3 Distribution of areas with changing land cover in the breeding area and the wintering area (a: in the breeding area, b: in the wintering area). there was a trend of habitat fragmentation. Although the landscape pattern improved after 2000, the landscape pattern was more fragmented in 2020 than in 2013 ( Figure S3).

(a) (b)
In the CNNR, the lake and arable land patches tended to be more complex in shape. The connectivity of all the landscapes in the CNNR was good, especially the lake, but the aggregation of the lake decreased slightly relative to the last century. After 2000, the connectivity and aggregation of drylands increased significantly, but TA B L E 4 Changes in the area of landscape in the CNNR (unit: km 2 ).

F I G U R E 4
Changes of landscape pattern in CNNR from 1980 to 2020(The percentages in parentheses in the subplots indicate the proportion of the changed/unchanged area to the total area of the CNNR. For example, from 1980 to 2010, the unchanged grassland area accounted for 20% of the CNNR, and the changed grassland area accounted for 13.91%).
in 2020 both connectivity and aggregation decreased than 2013 ( Figure S4). The index calculation and more detailed analysis results can be found in the Appendix S1.

| Correlation analyses between landscape pattern indices and the individual number of Blacknecked Crane
Habitat fragmentation existed in both breeding and wintering areas.
However, the number of Black-necked Cranes continued to increase, and habitat fragmentation did not significantly impact the number of   (Table S11,   Table S12).

| DISCUSS ION
The habitat fragmentation was higher in the wintering area of the Black-necked Crane. However, the habitat fragmentation trend ex-

| Analyses of the influencing factors of habitat landscape pattern change and landscape fragmentation at the large spatial scale
High landscape and habitat fragmentation in China are usually caused by anthropogenic disturbances, such as deforestation, agricultural reclamation, and urbanization (Dixo et al., 2009;Zhong et al., 2016). With the accelerated urbanization of breeding and wintering areas, the development of lakes for plateau development, the transformation of swamps, and the expansion of building sites and other human activities (Kong et al., 2018;Ruan et al., 2022) changed the landscape pattern of the Black-necked Crane habitats, and impact of human activities on the landscape pattern was more evident in the wintering area (Kong et al., 2011;Li, 1996;. The degradation and disappearance of wetland landscapes exacerbate the fragmentation of suitable habitats for Blacknecked Cranes. However, policies (e.g., reforestation and wetland retreat) can cause landscape patterns to be affected by human activities to different degrees at different times (Kong et al., 2011), which also determines the degree of interconversion and transformation of wetland landscapes with other different landscapes.
The spatial heterogeneity of the landscape on a large scale is usually determined by climate and topography, with a general increase in breeding and wintering open water areas, which may be related to the warming and humidification in the northwest. In the context of global climate change, the Tibetan Plateau experienced extreme changes . Rising temperatures led to massive glaciers and permafrost melting, increased river runoff, and rising lake levels (Tang et al., 2019). Since the mid-1980s, annual precipitation in the northwest has increased significantly, and temperature increases have accelerated (Gong et al., 2022;Shi et al., 2002;Zhang et al., 2019). The increased precipitation, higher temperatures, and the melting of glaciers and snowpack brought more runoff to rivers in the northwest, resulting in an increase in the areas of most lakes over the past 30 years (Fang et al., 2018). The results of the study also confirmed that the lake areas in both the breeding and wintering areas of Black-necked Cranes tended to increase over the past 30 years.

| Analyses of the influencing factors of habitat landscape pattern change and landscape fragmentation at the small spatial scale
Changes in the landscape pattern of the Caohai region have been strongly influenced by human activities. Especially before 1980, the natural landscape, such as wetlands in the Caohai area, was reduced due to over-exploitation of agriculture and other human activities (Ran et al., 2017;Ru et al., 2019), which seriously affected the habitat of Black-necked Cranes. The restoration and maintenance of the wetland landscape in the Caohai region benefited from the establishment of the Caohai National Nature Reserve after 1980, the implementation of the policy of returning farmland to wetlands and forests, and the increase in people's environmental awareness (Geng & Song, 1990;, and the results of the study also indicated that the landscape connectivity and aggregation in the Caohai region have been in a good and relatively stable state since 1980. However, the expansion of the scale of building land in recent years reflects that the impact of human activities in the Caohai area is gradually deepening (Zhao & Yang, 2021). By keeping human activities within the carrying capacity of the reserve, the landscape pattern and ecology of Caohai can be continuously protected, thus ensuring the survival space of wintering Black-necked Cranes.

| Analyses of influencing factors for the correlation between landscape and the individual number of black-necked cranes
The conversion of natural habitats to agricultural land was a fundamental cause of biodiversity loss (Tscharntke et al., 2005). However, a growing number of studies indicated that waterbirds such as Blacknecked Cranes might benefit from cultivated landscapes (Sundar & Subramanya, 2010). Black-necked Crane subpopulations wintering in the Yarlung Zangbo River basin in Tibet, China, and the Dashanbao National Nature Reserve in Yunnan Province show a selection for farmland habitats (Tsamchu & Bishop, 2005). In the Caohai, Guizhou Province, arable land gradually becomes the main foraging site for wintering Black-necked Cranes Wu et al., 2021).

Studies by scholars in India and Africa have shown that Blue
Cranes use agricultural fields and pastures (Allan, 1995), and Sarus Cranes and Brolgas also use treeless agricultural landscapes scattered in wetlands (Mukherjee et al., 2002;Sheldon, 2005;Sundar & Kittur, 2012). Breeding populations of woolly necked Storks are surprisingly high in the intensively cultivated and crowded areas of Jhajjar and Rohtak, India (Kittur & Sundar, 2021). The Black-headed Ibis still maintain habitats in highly disturbed urban areas (Chaudhury & Koli, 2018;Koli et al., 2019;Sundar, 2006).
A possible reason for this phenomenon of closer spatial association of waterbird foraging sites with agricultural landscapes is that the shrinking and fragmentation of suitable habitat and foraging sites has led to the shrinking of food sources in natural landscapes, while food sources in agricultural landscapes support the survival of waterbirds such as the Black-necked Crane Li, 1999;Yang et al., 2018), making the use of landscape patches by waterbirds such as the Black-necked Crane has been altered.
Although the conservation of Black-necked cranes should focus on cultivated landscapes, the wetland provides security and isolation from human disturbance (Li et al., 2022) and are an important basis for the survival of the Black-necked cranes and other waterbirds. For example, in Indian mosaic landscapes with both agricultural components and natural wetlands, many waterbird species prefer natural wetlands (Sundar, 2004(Sundar, , 2006 and even avoid paddy fields (Maeda, 2001). Some waterfowl use natural wetlands to paddy fields in almost all seasons, and flooded agricultural lands do not adequately compensate for the lack of natural wetlands (Sundar, 2006).

ACK N O WLE D G E M ENTS
Thanks to Dr. Xiaoran Lv for her suggestions and help in improving the manuscript. We are grateful to the Editors and Reviewers for all their constructive suggestions for improving this manuscript. This

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

DATA AVA I L A B I L I T Y S TAT E M E N T
The remote sensing data on land use in this study can be found in the