Integrating the spatial proximity effect into the assessment of changes in ecosystem services for biodiversity conservation

Ecological Indicators 70 (2016) 382–392

Contents lists available at ScienceDirect

Ecological Indicators journal homepage: www.elsevier.com/locate/ecolind

Integrating the spatial proximity effect into the assessment of changes in ecosystem services for biodiversity conservation Yanfang Liu a,b , Lei Zhang a,∗ , Xiaojian Wei c,d , Peng Xie a a

School of Resource and Environment Sciences, Wuhan University, 129 Luoyu Road, Wuhan 430079, China Collaborative Innovation Center of Geospatial information technology, Wuhan University, 129 Luoyu Road, Wuhan 430079, China Key Laboratory of Watershed Ecology and Geographical Environment Monitoring, National Administration of Surveying, Mapping and Geoinformation, China d School of Geomatics, East China University of Technology, 418 Guanglan Road, Nanchang 330013, China b c

a r t i c l e

i n f o

Article history: Received 8 September 2015 Received in revised form 9 June 2016 Accepted 13 June 2016 Keywords: Land cover change Proximity effect Ecosystem service Biodiversity conservation Land use planning

a b s t r a c t The assessment of the value of ecosystem services is a valuable tool for biodiversity conservation that can facilitate better environmental policy decision-making and land management, and can help land managers develop interventions to compensate for biodiversity loss at the patch level. Previous studies have suggested that it is appropriate to assess the value of biodiversity for conservation planning by considering both the condition of the landscape and the spatial configuration of adjacent land uses that can be reflected as a proximity effect. This research examines the influence of spatial proximity on biodiversity conservation from the ecosystem service perspective based on the assumption that the variation in the proximity effect caused by land cover change has positive or negative impacts on ecological services. Three factors related to the spatial characteristics of the landscape were considered in this approach: the relative artificiality of the land cover types, the distance decay effect of patches and the impact of one land cover type on others. The proximity effect change (PEC ) parameter reflected the relationship between the spatial proximity effect and biodiversity conservation. The results of a quantitative and spatial comparative analysis of the proposed method and the conventional method in Yingkou for the periods of 2000–2005 and 2005–2010 showed that the former can account for the temporal and spatial changes in ecosystem services for biodiversity conservation that were caused by patch-level changes as well as the interaction between the altered and adjacent patches from a spatial perspective. The metric can also identify the most critical areas for biodiversity protection and inform the efficient allocation of limited land resources for nature conservation to maximize the benefit to biodiversity by guiding the process of land-use change, particularly urbanization and agriculture. Future studies should focus on the other important factors that are applicable to the assessment of the value of biodiversity conservation in socio-ecological systems, where society and nature are mutually capable of fulfilling their roles. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Ecosystem services have been defined as the benefits that people obtain, either directly or indirectly, from various ecosystems (Costanza et al., 1997; MEA, 2005a,b), and they are provided at various spatial and temporal scales (Hein et al., 2006), which are generally accepted to be structurally and functionally complex, spatially variable, and temporally dynamic (Wu, 2013). Land use change can have beneficial or detrimental effects on ecosystem services (Li et al., 2014), especially urban expansion (Su et al.,

∗ Corresponding author. E-mail addresses: [email protected], [email protected] (L. Zhang). http://dx.doi.org/10.1016/j.ecolind.2016.06.019 1470-160X/© 2016 Elsevier Ltd. All rights reserved.

2014; Qiu et al., 2015) and agriculture (Li et al., 2014). Over the last decade, regional ecological services have been significantly degraded (Collin and Melloul, 2001; Lautenbach et al., 2011), and biodiversity, in particular, has deteriorated (Foley et al., 2005). To reverse this situation, numerous studies have considered the interactions between ecosystems and land use (Kreuter et al., 2001; Li et al., 2010a; Camacho-Valdez et al., 2014), and the assessment of ecosystem service value for conservation is regarded as a valuable tool to facilitate better environmental policy decision-making and environmental management (MEA, 2005a; Pejchar and Mooney, 2009; Camacho-Valdez et al., 2014; Wang et al., 2015). Biodiversity has many definitions and multiple measures. It can be defined as “the diversity of life on Earth” (MEA, 2005b), and it is always regarded as “a regulator of underpinning ecosystem

Y. Liu et al. / Ecological Indicators 70 (2016) 382–392

processes, as a final ecosystem service and as a good” (Mace et al., 2012). At the patch level, biodiversity is significantly influenced by land use, such as agricultural development (Swift et al., 2004). It is related to four factors that influence habitat quality: the relative impact of each threat, the relative sensitivity of each habitat to each threat, the distance between the sources of threats and the habitats, and the degree to which the land use is legally protected (Tallis et al., 2011), and threats are sometimes human-dominated landscapes, such as cropland and urban areas (Bai et al., 2011). In this study, biodiversity is regarded as an ecosystem service as defined by Costanza et al. (1997) and Xie et al. (2003). Many methods have been proposed to assess the value of ecosystem services (Costanza et al., 1997; Eigenbrod et al., 2010; Syrbe and Walz, 2012; Ng et al., 2013), and they are divided into two categories: primary data-based and biome- or LULC proxy-based (Su et al., 2014). The value of the ecosystem services delivered by each land cover category has been widely assessed (Costanza et al., 1997; Xie et al., 2003; Ng et al., 2013), but estimations that are based solely on the condition of the landscape may be inappropriate for conservation planning without considering spatial configuration, habitat quality, landscape structure or adjacent land uses (Tallis and Polasky, 2009; Frank et al., 2012; Baral et al., 2013; Baral et al., 2014). However, the spatial aspects of landscape heterogeneity and configuration play a significant role in the maintenance biodiversity (Syrbe and Walz, 2012) because the capacity for providing goods and services within an ecosystem is not homogeneous across landscapes, and ecosystem services are not static phenomena (Fisher et al., 2009; Ng et al., 2013). To adequately evaluate changes in biodiversity caused by land use change from an ecosystem service perspective, knowledge of both the spatial characteristics of the landscape and habitat condition is required as well as the relationship with the surrounding landscape (Baral et al., 2014). Landscape spatial characteristics are complicated and related to other ecologically significant variables that can result in undervaluing or overvaluing ecosystem services (Ng et al., 2013). Several studies have attempted to take landscape spatial characteristics into account when valuing ecosystem services for biodiversity conservation, such as landscape connectivity (Ng et al., 2013) and landscape structure (Frank et al., 2012). The proximity index, as one of the landscape spatial characteristics important to ecological conservation, defines the spatial context of a patch in relation to neighboring patches of the same type (Gustafson and Parker, 1992). The area and nearest-neighbor distance of a patch and its neighboring patches are always considered in the assessment (Gustafson and Parker, 1992). Many studies have considered proximity in the assessment of ecosystem service value (Tran et al., 2010), habitat quality related to biodiversity (Frank et al., 2012), habitat isolation (Su et al., 2010; Ng et al., 2011; Xun et al., 2014) and species richness (Houlahan and Findlay, 2003). Proximity effects caused by changes in land use, especially due to urban sprawl (Su et al., 2010), have been shown to influence ecological processes and the dynamics of local plant and animal populations or landscape qualities. Therefore, proximity should be considered to properly account for the spatial variability in ecosystem service values for biodiversity conservation caused by landscape configuration dynamics. Our study was organized around two main research questions: (1) How does a change in the proximity effect influence ecosystem services for biodiversity conservation, and what land use changes cause such a proximity effect? (2) How do the spatio-temporal changes in patches affect biodiversity conservation, and how can biodiversity be maintained in land-use planning? To accomplish the proposed objective, Yingkou, which has undergone significant habitat loss and land cover change due to rapid urbanization during the periods of 2000–2005 and 2005–2010, was selected as a case study. A metric, the proximity effect change (PEC ) of patches, was calculated and analyzed to explain the relationship between the

383

Fig. 1. Location of the study area.

proximity effect and biodiversity conservation. A comparative analysis of ecosystem services using the proposed method (ESV Bp c ), with considers the influence of the proximity effect, and the conventional method (ESV Bc ) was conducted to analyze temporal and spatial changes in biodiversity conservation. The proposed method can identify that changes in biodiversity are not only caused by changes in patch size, but they also result from the influence of the proximity effect considering some related factors, such as the relative artificiality of land cover types, the distance decay effect of patches, and the impact of one land cover type on others. The method can also identify key areas for conservation and efficiently allocate land for nature conservation to maximize the benefits to biodiversity. 2. Materials and methods 2.1. Study area The study was conducted in Yingkou City, which is located in the northwestern part of the Liaodong Peninsula in China (Fig. 1), and the entire area covers 527,900 ha and extends from 39◦ 55 –40◦ 56 N and 121◦ 56 –123◦ 02 E. This region is mountainous, and its eastern section is near Liaodongwan. Over the past 10 years, the area has experienced remarkably rapid economic growth and urbanization, with a population of 2.35 million and a gross domestic product of 100.24 billion yuan in 2010 (LPPG, 2011). In 2000–2010, the amount of urban area increased from 12.47% to 15.04% of the study area, and cropland decreased from 32.54% to 31.16% (Table 4). These land cover changes have significantly impacted ecosystem processes. 2.2. Data sources and preparation Land cover types were derived from Landsat Thematic Mapper imagery taken in 2000, 2005, and 2010 (resolution: 30 m × 30 m). First, an atmospheric correction was performed with ENVI5.1, and a topographic correction based on a topographic map was then

384

Y. Liu et al. / Ecological Indicators 70 (2016) 382–392

Table 1 Land cover classification. Land cover type

Code

Description

Cropland Forest Urban

1 2 3

Water Other land

4 5

Land cultivated for economic purposes with a high degree of human activities, e.g., fields and orchards Land covered by plants that are economically or ecologically protected Land intensively influenced by human activities with a hard surface and minimal vegetation, e.g., roads, residential land, commercial land, industrial land Land with surface water influenced by humans to a certain degree Land with minimal human intervention, e.g., grasslands and bare land

Fig. 2. Distribution of the landscape classifications.

used to correct the 2010 images using the second-order polynomial geometric model. An image-to-image registration based on the 2010 images was conducted, and the root-mean-square error was limited to within 0.5 pixels. Next, the eCognition objectoriented classification was used to segment the image after the false color composite; the decision tree was boiled to classify the images (Zhang et al., 2014); and a visual interpretation method was employed to modify the classification. Finally, a classification scheme consisting of five land cover types, namely, cropland, forest, urban, water and other land (Table 1 and Fig. 2), was used to implement the classification. A total of 1180 samples, both unchanged and changed, from ground survey data and stratified random sampling were selected during a field survey in 2009 to assess the accuracy of the classification. The overall accuracies were greater than 90%, and the Kappa coefficients were greater than 0.8. 2.3. Methods 2.3.1. Value assessment of biodiversity conservation affected by the proximity effect This study employed a methodology for calculating biodiversity conservation considering the proximity effect based on the unit value of ecosystem services for each land cover type proposed by Costanza et al. (1997) and Xie et al. (2003). The proposed method was based on the assumption that the influence of the change in the proximity effect due to the change in land cover has a positive or negative impact on the ecological service for biodiversity conservation. Assessing the value of biodiversity at the patch level can help land managers employ interventions to compensate for diversity loss and determine which ecosystem services are the most sensitive to biodiversity (Swift et al., 2004). Moreover, habitat quality is sometimes used as a proxy for biodiversity (Tallis et al., 2011), so the value of biodiversity conservation in patches should be assessed from this perspective. The spatial proximity effect as well as its

impact on biodiversity would not change if there is no change in land cover, so to better reflect how the changing trend in biodiversity conservation is influenced by the proximity effect, only the altered patches were considered in this study. The conventional method for measuring changes in the ecosystem service value for biodiversity conservation for all patches (ESV Bc ) employs the following formula: ESV− Bc =



ESV− Bck =



k

(V C  k − VCk ) × Ak

(1)

k

where VCk and VC´ık signify the value coefficients of the ecosystem services for biodiversity conservation of the altered patch, k, before and after the change, and Ak indicates the area of k. To assess the spatial dynamics of the biological conservation service, the changes in the ecosystem service value for biodiversity conservation of all patches considering the proximity effect (ESV Bp c ) is proposed to be calculated as follows: ESV− Bp

c

=

 k

ESV− Bp

ck

=



VCp

k

× (V C  k − VCk ) × Ak

(2)

k

where VCp k is the biodiversity conservation value coefficient of the changed patch, k, based on the proximity effect. 2.3.2. Assessment of the biodiversity conservation value coefficients (VCk ) Studies of ecosystem services have become popular in landscape ecology since the research of Costanza et al. (1997), so to appropriately assess ecosystem services in China, Xie et al. (2003) established a table of the equivalent value per unit area of Chinese terrestrial ecosystem services based on Costanza et al. (1997) by integrating the results of ecological surveys by professionals. In this study, the above table was used with minimal modifications (Table 2); for example, desert and grassland were classified as other land. The economic value of one equivalent value per unit

Y. Liu et al. / Ecological Indicators 70 (2016) 382–392

385

Table 2 The value coefficients of ecosystem services for biodiversity conservation. Biodiversity conservation

Equivalent value per unit area of Chinese terrestrial ecosystem services

The value coefficients of ecosystem services (Yuan ha−1 year−1 )

Forest Cropland Water Other land Urban

3.26 0.71 2.49 0.49 0

10222.87 2226.45 7808.26 1536.57 0

area was equal to 1/7 of the average yield of the market value of grain (Xie et al., 2003; Li et al., 2010a). The average actual food yield of cropland was 6864.27 kg hm−2 (LBS, 2001-2011), and the average price for grain was 3.20 yuan kg−1 in 2010 (LPPG, 2011). Therefore, the economic value of one equivalent value per unit for Yingkou was 3135.85 yuan hm−2 , and the values of the ecosystem service coefficients for biodiversity conservation are shown in Table 2. 2.3.3. Assessment of the biodiversity conservation value coefficients of changed patches based on the proximity effect (VCp ) The value coefficient for biodiversity conservation based on the proximity effect is represented by VCp , and its measurement consists of three elements (Fig. 3) that are also considered in the quantification of habitat quality as a proxy of biodiversity (Bai et al., 2011; Tallis et al., 2011). Due to the positive or negative impact of the proximity effect on biodiversity conservation and the positive or negative change in biodiversity conservation, the VCp should be calculated in two aspects as follows: VCp = {

PEC ,

PEC ≥ 1

1/PEC ,

PEC < 1

PEC = 1 − (PE  − PE) = 1 − (A r ×

(3)

 i

Di Ii − Ar ×



Dj Ij )

(4)

j

where PEC is the influence of the proximity effect on the change in biodiversity around the altered patch. It is calculated as the difference between the proximity effect related to the patch before (i.e., PE) and after (i.e., PE´ı) the change, and the calculation of the met-

ric, PE, is adapted from the proximity index (Gustafson and Parker, 1992) with different classes and different factors. If the proximity effect of the changed patch increases, there is a negative effect on biodiversity conservation (i.e., PEC is less than 1), but there is a positive effect when it decreases (i.e., PEC is greater than 1). The values of Ar and A´ır are the relative artificiality of the patch before and after the change; Di and Dj refer to the distance decay effect of the patch on the adjacent land cover types i and j before and after the change; Ii and Ij are the impacts between the patch and the adjacent patches before and after the change.

2.3.3.1. Assessment of the relative artificiality of land cover types. The relative degree of artificiality is defined as the degree of impact of human activities on the ecosystem (Bahamondes, 2003), and the highest relative degrees of artificiality of land cover types, which represent the highest influence by human activity, significantly impact biodiversity (MEA, 2005b). In early studies, Westhoff (1983) classified vegetation types into four systems: cultural, seminatural, sub-natural and natural. Ferrari et al. (2008) defined five degrees of artificiality: urbanized, cropland, semi-natural, subnatural and natural. Therefore, the five land cover types were arranged by their degrees of artificiality according to the abovementioned classification systems and the actual situation: urban, cropland, water, forest, and other land. The artificiality weights were 10, 5, 3, 2, and 1 (1 = lowest artificiality, 10 = highest artificiality), respectively, following Ferrarini and Tomaselli (2010) with slight modifications and according to the degrees of impact of human activities on the land cover types (Yue et al., 2004). The rela-

Fig. 3. An example illustrating the relationship between the proximity effect and the three elements of the value measurement. The changed patches I, II and their adjacent patches (1–7) represent the different land cover types, and the effect of proximity on patch I was different than that on patch II.

386

Y. Liu et al. / Ecological Indicators 70 (2016) 382–392

Table 3 The relative impact of one land cover on other land cover types. Land cover type

Land cover type

The relative impact

Urban Urban Urban Urban Cropland Cropland

Cropland Forest Water Other land Forest Water

0.367 0.165 0.147 0.098 0.114 0.109

Table 4 Land cover changes in 2000–2010. Land cover type

Cropland Forest Urban Water Other land

2000

2005

2010

Land cover change in 2000–2010

Area (ha)

%

Area (ha)

%

Area (ha)

%

Area (ha)

%

171755.82 245132.55 65808.19 35934.46 9272.82

32.54 46.44 12.47 6.81 1.76

168125.45 244089.40 70114.43 36360.42 9214.14

31.85 46.24 13.28 6.89 1.75

164512.16 243814.84 79410.56 30491.36 9674.92

31.16 46.19 15.04 5.78 1.83

−7243.66 −1317.71 13602.37 −5443.10 402.09

−1.37 −0.25 2.58 −1.03 0.08

which the buffer characteristic exerts an influence, which can be expressed as follows:

Di =



e−dk

i /d

× Ak i /Asum

(6)

k

where Ak i and Asum are the area of the adjacent patch k of land cover i and the sum of the areas of the adjacent patches, respectively; dk i is the distance between the changed patch and the patch k; and d is the distance threshold. The range of the proximity effect on vegetation biomass, species, and matrix type has been investigated in many studies (Gustafson and Parker, 1992; Dauber et al., 2003; Tran et al., 2010; Sung, 2015). For example, Sung (2015) concluded that the threshold distance was 250 m for the cranes that are widely distributed in Yingkou. Gustafson and Parker (1992) used the distance of 300 m in the proximity index to explore the relationships between the proportion of a land cover type and indices of landscape spatial pattern. In this study, 300 m was adopted based on expert experience and the previous literature without considering environmental variables, specific species and other related factors. Fig. 4. Spatial distribution of land cover change, e.g., 21 means that forest changed into cropland at the given time. See Table 1 for an explanation of the codes.

tive artificiality of land cover types was extracted from the relative artificiality of the vegetation (Ferrari et al., 2008) as follows: Ar = A/Amax ,

(5)

where Ar is the relative artificiality of the land cover type; A is the artificiality weight of the land cover type; and Amax is the maximum artificiality weight.

2.3.3.2. Assessment of the distance decay effect of the patch on adjacent land cover types. The influence of proximity with increasing distance can be expressed by a distance decay function that considers area and distance (Levin et al., 2007) because the impact of the large, closed patches on the center patch was significantly greater than that of the small remote patches (Tischendorf et al., 2003). The distance decay function is adopted to reflect the distance over

2.3.3.3. Assessment of the relative impact of one land cover type on others. The relative impact of one land cover on others defines the relative ecological impact between land cover types, and a high value indicates a significant influence on ecosystem process. For example, the influence of urban land cover on water is more significant than the influence of cropland on water (Du et al., 2010). This study primarily considers the influences of an urban area or cropland on other land cover types (Table 3), and the impact of one land cover on others is determined by the Delphi method (Li et al., 2010b). In the Delphi process, 30 experts from universities, research institutes, governments and government agencies in Yingkou and other relevant institutions participated on a panel. Questionnaires related to the value scores of the impact between land cover types were anonymously sent to these experts by email, and the statistical results of questionnaires were returned to the experts who could modify their responses according to the results. Finally, the weights were determined after several rounds of anonymous consultation and feedback.

Y. Liu et al. / Ecological Indicators 70 (2016) 382–392

387

Table 5 Scope and rank ratio of the influence of the proximity effect change on biodiversity (PEC ). Rank of PEC

PEC 1 PEC 2 PEC 3 PEC 4 PEC 5

Scope of PEC rank

higher high medium low lower

1.00–1.24 0.97–1.00 0.89–0.97 0.83–0.89 0.69–0.83

Rank ratio of PEC in number of change patches 2000–2005

2005–2010

13.07 25.02 19.35 28.46 14.10

30.90 11.26 19.88 18.32 19.64

Fig. 5. Influence of the variation in proximity on ecosystem services (PEC ) in 2000–2005 and 2005–2010. (a) Characteristic degrees of PEC in 2000–2005. (b) Characteristic degrees of PEC in 2005–2010. (c) Characteristic PEC in 2000–2005 by changed land cover types. (d) Characteristic of PEC in 2005–2010 by changed land cover types.

3. Results 3.1. Results of land cover change

the proximity effect increased after the landscape changed, which negatively affected biodiversity. The ranks of the ratios of the number of changed patches significantly differed, and the number of changed patches differed

Land cover change is a key factor that drives changes in ecosystem services (MEA, 2005a), and land cover in Yingkou substantially changed from 2000 to 2010 (Table 4). In 2000–2010, urban land cover increased from 12.47% to 15.04%, while cropland decreased from 32.54% to 31.16%. Further analyses of land cover change were performed to reveal the trends in the conversion and spatial changes among land cover types (Fig. 4). The altered patches were mostly distributed in the north and southwest of the city within the two periods, 2000–2005 and 2005–2010, and cropland and forest were mostly transformed into urban in the west of the city. Meanwhile, urban land cover was hardly converted into other types. 3.2. Characteristics of the influence of the proximity effect change (PEC ) on biodiversity conservation To easily identify the spatial differences in proximity effects, the values of PEC were first divided into five categories: higher, high, medium, low, and lower (Table 5). PEC1 indicated that the proximity effect decreased after the patch changed, and these changes positively affected biodiversity, including the conversion of urban land cover into others and most of cropland into forest, water, and other land (Fig. 5). For the other categories besides PEC equal to 1,

Fig. 6. Spatial distribution of the variation in the influence of proximity on ecosystem services (PEC ) in 2000–2005 and 2005–2010.

388

Y. Liu et al. / Ecological Indicators 70 (2016) 382–392

minimally during the two periods (Table 5). The ratio of PEC 1 increased markedly by approximately 27.83%; in contrast, the ratio of PEC4 and PEC 5 slightly decreased by approximately 4.60%. This finding demonstrated that the land-use policy in Yingkou was reasonable from the ecological perspective for the last 10 years. The spatial characteristics of PEC for the two study periods are shown in Fig. 6. The changed patches with a strong proximity negative effect on biodiversity were mainly scattered in the west of the city, which was related to the distribution of urban expansion. In contrast, the patches with strong positive proximity effects were mainly located in the southeast of the city, which was similar to the distribution of forest. 3.3. Quantitative characteristics of the changes in biodiversity conservation using the proposed method (ESV Bp c ) and the conventional method (ESV Bc ) The quantitative characteristics of the proposed method and the conventional method are shown for 2000–2005 and 2005–2010 (Figs. 7 and 8). The distribution of all of the points in Fig. 7 shows that the value of ESV Bp c increased as biodiversity conservation increased (greater than zero) and decreased as biodiversity conservation decreased (less than zero) compared with the value of ESV Bc . The trends in the changes of each transformed land type between ESV Bp c and ESV Bc were similar (Fig. 8). For example, the change from other land to forest resulted in a significant increase in the value of biodiversity conservation. The comparative analysis of the total ESV Bp c and ESV Bc values is shown in Fig. 9. ESV Bp c over ESV Bc decreased to 17.45% and 12.07% in 2000–2005 and 2005–2010, respectively. Among the transformed land cover types, most changed except for the conversion of other land to water. 3.4. Spatial characteristics of the changes for biodiversity conservation using the proposed (ESV Bp c ) and conventional methods (ESV Bc ) 3.4.1. Decreasing biodiversity conservation The spatial characteristics of the changes in terms of decreasing biodiversity conservation in two periods are shown in Fig. 10. Patches with low or a lower degree had a strong negative proximity effect on the ecosystem services scattered in the north. As shown in Fig. 10e, cropland converted to urban land cover with unchanged adjacent patches dropped by one level because the proximity effect between urban and adjacent patches was larger than that of cropland. As shown in Fig. 10f, forest or cropland with the same adjacent land cover types and similar areas were converted into urban cover; the value of forest was different because it had a different proximity effect on biodiversity. Fig. 10h shows two water patches with similar areas and different adjacent patches; the negative influence of the change in the proximity of the upper patch on ecosystem services was obvious due to the different landscape structure. 3.4.2. Increasing biodiversity conservation The spatial characteristics of ESV Bc in terms of increasing biodiversity in 2000–2005 and 2005–2010 are shown in Fig. 11. The degrees of ESV Bc and ESV Bp c of most patches located in the northwest of the region were lower during both periods; this result indicated that land cover change in this area had a minimal positive proximity effect on ecosystem services. The patches that were converted from cropland to water and from urban cover to cropland or forest underwent obvious changes according to the comparative analysis. These changes were mostly observed around cropland, forest and water, which provided greater ecosystem services for biodiversity conservation.

4. Discussion 4.1. Land cover change, proximity effect and the impact on biodiversity Much of the previous literature has considered the interaction between biodiversity and land use (Sala et al., 2000; Haines-Young, 2009). Land cover changes are the major factors driving the change in biodiversity, and the impact on biodiversity is likely to be more significant than due to climate change, nitrogen deposition and other factors (Sala et al., 2000; MEA, 2005b; Haines-Young, 2009). Although the impact of land use change may be of two types, beneficial or detrimental, land cover change has a negative effect in most regions (Haines-Young, 2009) due to the loss and fragmentation of habitat (Ng et al., 2011) and landscape characteristics (Ng et al., 2013), especially the transformation to urban cover (Su et al., 2014) and agriculture (Green et al., 2005). The proximity effect is an important landscape characteristic in conservation planning, and adjacent landscapes play a significant role in biodiversity conservation. Many studies have examined the impacts of land use on adjacent habitats that would result in habitat fragmentation and species loss, such as adjacent land use (Houlahan and Findlay, 2003), transportation corridors (Houlahan and Findlay, 2003; Hardt et al., 2013) and the proximity of urbanization and agriculture (Tran et al., 2010). In the present study, an empirical method was proposed to demonstrate how the proximity effect caused by land cover change influences biodiversity conservation. A high PEC value represents there have an obvious positive proximity effect on biological conservation, so the spatial allocation of changed patches in a location is reasonable from an ecological perspective. On the other hand, a patch with a low value would have a negative effect; in particular, the PEC value of a patch of other land converted to urban land is close to 1, which indicates that urban expansion has a minimal negative proximity effect on ecosystem services.

4.2. The spatial changes in biodiversity conservation using a comparative analysis Spatial considerations are important for the estimation and valuation of ecosystem service because it is necessary that the loss and degradation of habitats be avoided for biodiversity conservation (Syrbe and Walz, 2012). Therefore, the method of assessment considered both the impact of landscape spatial characteristics and the spatial interaction between patches on the ecosystem service (Ng et al., 2013) and the analysis of the spatial changes in ecosystem services. As illustrated by the case study, the result of the spatial characteristics analysis indicates that changes in patches with the same area at different locations may alter ecosystem services for biodiversity conservation differently due to the different landscape structures. Taking the decrease in biodiversity as an example, when two patches have different land cover types, similar areas, the same adjacent land cover types, and are converted into the same landscape type or when two patches have the same landscape type, similar areas, different adjacent patches, and are converted into the same landscape type, ESV Bc and ESV Bp c exhibit obvious differences that are mainly caused by the differences in landscape structure (Fig. 10). On the other hand, from the perspective of increasing biodiversity, ESV Bp c will be higher than ESV Bc when adjacent patches can provide greater ecosystem services (Fig. 11). Therefore, one of the improvements of the proposed metric is the capability to develop a method for assessing the change in biodiversity conservation associated with the features of the patches and the relationship between the changed patches and the adjacent patches.

Y. Liu et al. / Ecological Indicators 70 (2016) 382–392

389

Fig. 7. Values of the conventional method (ESV Bc ) and the proposed method (ESV Bp c ) for landscape changes in 2000–2005 (a) and 2005–2010 (b). In 2005–2010, the lowest value of one patch was not displayed because it exceeded the display range.

Fig. 8. Values of the conventional method (ESV Bc ) and the proposed method (ESV Bp c ) for different changed landscape types in 2000–2005 (a) and 2005–2010 (b). In 2005–2010, the lowest value of one patch was not displayed because it exceeded the display range.

Fig. 9. Values of the conventional method (ESV Bc ) and the proposed method (ESV Bp c ) for all patches in 2000–2005 and 2005–2010.

4.3. Application of the metric in land-use planning Although several studies have examined ecosystem services (Costanza et al., 1997; Fisher et al., 2009; Li et al., 2010a; Wang et al., 2015), incorporating them into land use planning poses many challenges (MEA, 2005a; Fisher et al., 2009; De Groot et al., 2010; Ng et al., 2013). Similar to the research of Ng et al. (2013), the method

proposed in this study is able to identify priority areas for land management for ecological protection in terms of both habitat size and landscape characteristics. In this study, changes to a patch with a low PEC value should be controlled, or the patch should be replaced in another location to decrease the negative effect on biodiversity conservation. Therefore, incorporating the PEC value into landscape planning will aid decision-making regarding the optimal allocation

390

Y. Liu et al. / Ecological Indicators 70 (2016) 382–392

Fig. 10. Comparison of the conventional method (ESV Bc ) and the proposed method (ESV Bp c ) for changed patches in terms of decreasing biodiversity; a) ESV Bc in 2000–2005; b) ESV Bp c in 2000–2005; c) ESV Bc in 2005–2010; d) ESV Bp c in 2005–2010. Patches marked within rectangles indicate the different degrees estimated by the two methods, but some patches were not marked. The base map in e, f, g, and h shows land cover types.

Fig. 11. Comparison of the conventional method (ESV Bc ) and the proposed method (ESV Bp c ) for changed patches in terms of increasing biodiversity; a) ESV Bc in 2000–2005; b) ESV Bp c in 2000–2005; c) ESV Bc in 2005–2010; d) ESV Bp c in 2005–2010. Patches marked within rectangles indicate the different degrees estimated by the two methods, but some patches were not marked. The base map in e, f, g, and h shows land cover types.

Y. Liu et al. / Ecological Indicators 70 (2016) 382–392

of land and biological conservation. The comparative analysis can inform how to minimize negative effects from changing patches in landscape planning, particularly urban expansion and cropland conversion. For example, the patch that changed to urban cover resulted in the landscape being more isolated (Fig. 9) and may have a significant negative impact on biodiversity (Tischendorf et al., 2003). Therefore, changing this patch may be unreasonable from an ecological perspective, so it should be replaced to improve land-use planning and ecological benefits.

391

accounting for biological conservation and regional development in landscape planning. In conclusion, this study provides a means to consider the spatial characteristics related to the influence of proximity in evaluating changes in ecosystem services for biodiversity conservation. It also provides information to minimize the negative effects of changing land cover, particularly urban expansion and cropland conversion.

Acknowledgments 4.4. Limitations and further studies Although the proposed methodology had successfully integrated landscape characteristics into the assessment of biodiversity conservation, there are still several limitations. On one hand, the parameters were adopted from previous studies and expert experiences due to the lack of appropriate data. For example, the coefficients for biodiversity conservation were borrowed from Xie et al. (2003) and the relative artificiality of land cover types determined by expert experience, so they could be valid for other regions with minor adjustments. In particular, the distance decay effect of a patch on adjacent land cover types was calculated using the same distance threshold, but it should be specific to the features of the land cover types, the dispersal distance threshold of the organisms and other ecologically important variables, to more effectively relate this element to ecological processes. However, this threshold is complicated and difficult to determine in practice and varies across different species. This issue should be considered as much as possible in future research. On the other hand, this method is only focused on the assessment of the value of ecosystem services from the ecological perspective and does not consider the economic development and social welfare. Therefore, further research of specific scenarios can introduce a participatory approach into the calculation of parameters to integrate the thoughts of local residents, environmental organizations, government workers or experts who are familiar with regional conditions (Jianzhong et al., 2006; Berbés-Blázquez, 2012; Paudyal et al., 2015; Van Oort et al., 2015; Zarandian et al., 2016). This improvement can identify both biophysical and socioeconomic value associate with ecological process in most regions, even in developing and data-poor countries with a lack of data or technical expertise. 5. Conclusion In this study, a place-based method was provided to introduce the influence of the proximity effect into the calculation of the change in biodiversity conservation from an ecosystem services perspective. Although it has many limitations, we believe that this study has made a significant contribution to the assessment of the spatial change in biodiversity conservation. To demonstrate the superiority of the proposed method for biological protection, a spatial analysis of the value coefficients for biodiversity conservation based on the proximity effect and a comparative assessment of the conventional method and the proposed method was conducted. The results of the empirical research in Yingkou in 2000–2005 and 2005–2010 demonstrated that the proposed method has the capability to identify the positive and negative impacts of the proximity effect caused by changes to patches on biodiversity conservation. In some situations, the proposed method identified the temporal and spatial changes in ecosystem services for biodiversity conservation that were not only caused by habitat size but also by the dynamics of the landscape structure related to the proximity effect of the changed patches from a spatial perspective. The method also provides information to identify the most critical areas for biodiversity protection and to guide the reasonable allocation of land by

This research was funded by Special Fund of Ministry of Land and Resources of China in the Public Interest (ID. 201511001). We sincerely thank to the editor for his guidance and patience in this paper. We would give our sincerely thank to anonymous reviewer for their constructive comments on the manuscript.

References Bahamondes, M., 2003. Poverty-environment patterns in a growing economy: farming communities in Arid Central Chile, 1991–99. World Dev. 31 (11), 1947–1957. Bai, Y., Zhuang, C., Ouyang, Z., Zheng, H., Jiang, B., 2011. Spatial characteristics between biodiversity and ecosystem services in a human-dominated watershed. Ecol. Complex. 8 (2), 177–183. Baral, H., Keenan, R.J., Fox, J.C., Stork, N.E., Kasel, S., 2013. Spatial assessment of ecosystem goods and services in complex production landscapes: a case study from south-eastern Australia. Ecol. Complex. 13, 35–45. Baral, H., Keenan, R.J., Sharma, S.K., Stork, N.E., Kasel, S., 2014. Spatial assessment and mapping of biodiversity and conservation priorities in a heavily modified and fragmented production landscape in north-central Victoria, Australia. Ecol. Indic. 36, 552–562. Berbés-Blázquez, M., 2012. A participatory assessment of ecosystem services and human wellbeing in rural Costa Rica using photo-voice. Environ. Manag. 49 (4), 862–875. Camacho-Valdez, V., Ruiz-Luna, A., Ghermandi, A., Berlanga-Robles, C.A., Nunes, P.A.L.D., 2014. Effects of land use changes on the ecosystem service values of coastal wetlands. Environ. Manag. 54 (4), 852–864. Collin, M.L., Melloul, A.J., 2001. Combined land-use and environmental factors for sustainable groundwater management. Urban Water 3 (3), 229–237. Costanza, R., d’Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S., O’Neill, R.V., Paruelo, J., Raskin, R.G., Sutton, P., van den Belt, M., 1997. The value of the world’s ecosystem services and natural capital. Nature 387 (6630), 253–260. Dauber, J., Hirsch, M., Simmering, D., Waldhardt, R., Otte, A., Wolters, V., 2003. Landscape structure as an indicator of biodiversity: matrix effects on species richness. Agric. Ecosyst. Environ. 98 (1–3), 321–329. De Groot, R.S., Alkemade, R., Braat, L., Hein, L., Willemen, L., 2010. Challenges in integrating the concept of ecosystem services and values in landscape planning, management and decision making. Ecol. Complex. 7 (3), 260–272. Du, N., Ottens, H., Sliuzas, R., 2010. Spatial impact of urban expansion on surface water bodies—A case study of Wuhan, China. Landsc. Urban Plan. 94 (3–4), 175–185. Eigenbrod, F., Armsworth, P.R., Anderson, B.J., Heinemeyer, A., Gillings, S., Roy, D.B., Thomas, C.D., Gaston, K.J., 2010. The impact of proxy-based methods on mapping the distribution of ecosystem services. J. Appl. Ecol. 47 (2), 377–385. Ferrari, C., Pezzi, G., Diani, L., Corazza, M., 2008. Evaluating landscape quality with vegetation naturalness maps: an index and some inferences. Appl. Veg. Sci. 11 (2), 243–250. Ferrarini, A., Tomaselli, M., 2010. A new approach to the analysis of adjacencies: potentials for landscape insights. Ecol. Model. 221 (16), 1889–1896. Fisher, B., Turner, R.K., Morling, P., 2009. Defining and classifying ecosystem services for decision making. Ecol. Econ. 68 (3), 643–653. Foley, J.A., DeFries, R., Asner, G.P., Barford, C., Bonan, G., Carpenter, S.R., Chapin, F.S., Coe, M.T., Daily, G.C., Gibbs, H.K., Helkowski, J.H., Holloway, T., Howard, E.A., Kucharik, C.J., Monfreda, C., Patz, J.A., Prentice, I.C., Ramankutty, N., Snyder, P.K., 2005. Global consequences of land use. Science 309 (5734), 570–574. Frank, S., Fürst, C., Koschke, L., Makeschin, F., 2012. A contribution towards a transfer of the ecosystem service concept to landscape planning using landscape metrics. Ecol. Indic. 21, 30–38. Green, R.E., Cornell, S.J., Scharlemann, J., Balmford, A., 2005. Farming and the fate of wild nature. Science 307 (5709), 550–555. Gustafson, E.J., Parker, G.R., 1992. Relationships between landcover proportion and indexes of landscape spatial parrern. Landsc. Ecol. 7 (2), 101–110. Haines-Young, R., 2009. Land use and biodiversity relationships. Land Use Policy 26, S178–S186. Hardt, E., Pereira-Silva, E.F.L., Dos Santos, R.F., Tamashiro, J.Y., Ragazzi, S., Lins, D.B.D.S., 2013. The influence of natural and anthropogenic landscapes on edge effects. Landsc. Urban Plan. 120, 59–69.

392

Y. Liu et al. / Ecological Indicators 70 (2016) 382–392

Hein, L., van Koppen, K., de Groot, R.S., van Ierland, E.C., 2006. Spatial scales, stakeholders and the valuation of ecosystem services. Ecol. Econ. 57 (2), 209–228. Houlahan, J.E., Findlay, C.S., 2003. The effects of adjacent land use on wetland amphibian species richness and community composition. Can. J. Fish. Aquat. Sci. 60 (9), 1078–1094. Jianzhong, Y., Yili, Z., Linshan, L., Wanqi, B., Huiyi, Z., Yulin, S., Du, Z., 2006. Residents’ perspectives and responses to environmental degradation in the upper Dadu River, eastern Tibetan Plateau. J. Geogr. Sci. 16, 293–305. Kreuter, U.P., Harris, H.G., Matlock, M.D., Lacey, R.E., 2001. Change in ecosystem service values in the San Antonio area, Texas. Ecol. Econ. 39 (3), 333–346. LBS, 2001–2011. Liaoning Statistical Yearbook China Statistics Press, Beijing, China. LPPG, 2011. Liaoning Yearbook 2011. China Statistical Press, Beijing, China. Lautenbach, S., Kugel, C., Lausch, A., Seppelt, R., 2011. Analysis of historic changes in regional ecosystem service provisioning using land use data. Ecol. Indic. 11 (2), 676–687. Levin, N., Lahav, H., Ramon, U., Heller, A., Nizry, G., Tsoar, A., Sagi, Y., 2007. Landscape continuity analysis: a new approach to conservation planning in Israel. Landsc. Urban Plan. 79 (1), 53–64. Li, T., Li, W., Qian, Z., 2010a. Variations in ecosystem service value in response to land use changes in Shenzhen. Ecol. Econ. 69 (7), 1427–1435. Li, Y., Sun, X., Zhu, X., Cao, H., 2010b. An early warning method of landscape ecological security in rapid urbanizing coastal areas and its application in Xiamen, China. Ecol. Model. 221 (19), 2251–2260. Li, F., Wang, R., Hu, D., Ye, Y., Yang, W., Liu, H., 2014. Measurement methods and applications for beneficial and detrimental effects of ecological services. Ecol. Indic. 47, 102–111. MEA, 2005a. Ecosystems and Human Well-being: Synthesis, 39–48. Island Press, Washington, DC, pp. 1–5. MEA, 2005b. Ecosystems and Human Well-being: Biodiversity Synthesis. World Resources Institute, Washington, DC, pp. 1–2. Mace, G.M., Norris, K., Fitter, A.H., 2012. Biodiversity and ecosystem services: a multilayered relationship. Trends Ecol. Evol. 27 (1), 19–26. Ng, C.N., Xie, Y.J., Yu, X.J., 2011. Measuring the spatio-temporal variation of habitat isolation due to rapid urbanization: a case study of the Shenzhen River cross-boundary catchment, China. Landsc. Urban Plan. 103 (1), 44–54. Ng, C.N., Xie, Y.J., Yu, X.J., 2013. Integrating landscape connectivity into the evaluation of ecosystem services for biodiversity conservation and its implications for landscape planning. Appl. Geogr. 42, 1–12. Paudyal, K., Baral, H., Burkhard, B., Bhandari, S.P., Keenan, R.J., 2015. Participatory assessment and mapping of ecosystem services in a data-poor region: Case study of community-managed forests in central Nepal. Ecosyst. Serv. 13, 81–92. Pejchar, L., Mooney, H.A., 2009. Invasive species, ecosystem services and human well-being. Trends Ecol. Evol. 24 (9), 497–504. Qiu, B., Li, H., Zhou, M., Zhang, L., 2015. Vulnerability of ecosystem services provisioning to urbanization: a case of China. Ecol. Indic. 57, 505–513. Sala, O.E., Chapin, F.S., Armesto, J.J., Berlow, E., Bloomfield, J., Dirzo, R., Huber-Sanwald, E., Huenneke, L.F., Jackson, R.B., Kinzig, A., Leemans, R., Lodge, D.M., Mooney, H.A., Oesterheld, M., Poff, N.L., Sykes, M.T., Walker, B.H., Walker, M., Wall, D.H., 2000. Biodiversity – Global biodiversity scenarios for the year 2100. Science 287 (5459), 1770–1774. Su, W., Gu, C., Yang, G., Chen, S., Zhen, F., 2010. Measuring the impact of urban sprawl on natural landscape pattern of the Western Taihu Lake watershed, China. Landsc. Urban Plan. 95 (1–2), 61–67. Su, S., Li, D., Hu, Y.N., Xiao, R., Zhang, Y., 2014. Spatially non-stationary response of ecosystem service value changes to urbanization in Shanghai, China. Ecol. Indic. 45, 332–339.

Sung, C.Y., 2015. Simulation of crane habitat fragmentation in the North and South Korean border region after Korean reunification. Landsc. Urban Plan. 134, 10–18. Swift, M.J., Izac, A.M.N., van Noordwijk, M., 2004. Biodiversity and ecosystem services in agricultural landscapes—are we asking the right questions? Agric. Ecosyst. Environ. 104 (1), 113–134. Syrbe, R., Walz, U., 2012. Spatial indicators for the assessment of ecosystem services: providing, benefiting and connecting areas and landscape metrics. Ecol. Indic. 21, 80–88. Tallis, H., Polasky, S., 2009. Mapping and valuing ecosystem services as an approach for conservation and natural-resource management. Ann. N. Y. Acad. Sci. 1162 (1), 265–283. Tallis, H.T., Ricketts, T., Guerry, A.D., Nelson, E., Ennaanay, D., Wolny, S., Olwero, N., Vigerstol, K., Pennington, D., Mendoza, G., Aukema, J., Foster, J., Forrest, J., Cameron, D., Lonsdorf, E., Kennedy, C., Verutes, G., Kim, C.K., Guannel, G., Papenfus, M., Toft, J., Marsik, M., Bernhardt, J., 2011. InVEST 2.0 Beta User’s Guide. The Natural Capital Project, Stanford. Tischendorf, L., Bender, D.J., Fahrig, L., 2003. Evaluation of patch isolation metrics in mosaic landscapes for specialist vs. generalist dispersers. Landsc. Ecol. 18 (1), 41–50. Tran, C.P., Bode, R.W., Smith, A.J., Kleppel, G.S., 2010. Land-use proximity as a basis for assessing stream water quality in New York State (USA). Ecol. Indic. 10 (3), 727–733. Van Oort, B., Bhatta, L.D., Baral, H., Rai, R.K., Dhakal, M., Rucevska, I., Adhikari, R., 2015. Assessing community values to support mapping of ecosystem services in the Koshi river basin, NepalAssessing community values to support mapping of ecosystem services in the Koshi river basin, Nepal. Ecosyst. Serv. 13, 70–80. Wang, J., Yan, S., Guo, Y., Li, J., Sun, G., 2015. The effects of land consolidation on the ecological connectivity based on ecosystem service value: a case study of Da’an land consolidation project in Jilin province. J. Geogr. Sci. 25 (5), 603–616. Westhoff, V., 1983. Man’s attitude towards vegetation. In: Holzner, W., Werger, M.J.A., Ikusima, I. (Eds.), Man’s Impact on Vegetation. Junk, Den Haag, NL, pp. 7–24. Wu, J., 2013. Landscape sustainability science: ecosystem services and human well-being in changing landscapes. Landsc. Ecol. 28 (6), 999–1023. Xie, G., Lu, C., Leng, Y., Zhen, D., Li, S., Abade, L., 2003. Ecological assets valuation of the Tibetan Plateau. J. Nat. Resour. 18 (2), 189–196. Xun, B., Yu, D., Liu, Y., Hao, R., Sun, Y., 2014. Quantifying isolation effect of urban growth on key ecological areas. Ecol. Eng. 69, 46–54. Yue, T.X., Xu, B., Liu, J.Y., 2004. A patch connectivity index and its change in relation to new wetland at the Yellow River Delta. Int. J. Remote Sens. 25 (21), 4617–4628. Zarandian, A., Baral, H., Yavari, A., Jafari, H., Stork, N., Ling, M., Amirnejad, H., 2016. Anthropogenic decline of ecosystem services threatens the integrity of the unique hyrcanian (Caspian) forests in Northern Iran. Forests 7 (3), 51. Zhang, Z.J., Li, A.N., Lei, G.B., Bian, J.H., Wu, B.F., 2014. Change detection of remote sensing images based on multiscale segmentation and decision tree algorithm over mountainous area: a case study in Panxi region, Sichuan Province. Acta Ecol. Sin. 34 (24), 7222–7232.