Habitat Connectivity Analysis for Township of Langley, BC

Urbanization has been identified as a significant threat to biodiversity, but little is known about the functional habitat connectivity of the Coastal Douglas-fir (CDF) Biogeoclimatic Classification Ecosystem (BEC) zone and the Western Hemlock (CWH) BEC Zone in urban areas. This study aimed to assess the functional habitat connectivity of the Township of Langley in British Columbia, Canada, to facilitate urban planning and urban biodiversity conservation. Using lidar-derived and land cover data, quantified habitat connectivity was quantified with the probability of connectivity index (PC) for three species (the western red-backed salamander, the American red squirrel, and the brown creeper) with different habitat requirements and mobility. Suitable habitat patches were filtered and selected based on the habitat requirements of each species, and habitat networks were generated considering dispersal distance to model the connectivity index in the Conefor software. Key patches, hub patches, and key-hub patches were identified based on the patch importance level generated from Conefor. The study found a large variance in PC and distribution of important patches among species in the landscape. The major connectivity corridor, identified according to the distribution of key-hub patches for all species, showed some important areas for overall connectivity improvement. Despite limitations on timeliness, computational power, and consideration of dispersal barriers, this study enhances understanding of local-scale connectivity and offers guidance for urban planners and developers seeking to incorporate biodiversity conservation in decision-making. Identification of key patches, hub patches, and key-hub patches also highlights potential areas for habitat restoration and connectivity improvement, which is particularly relevant for urban green infrastructure design that supports both human well-being and biodiversity conservation.

To facilitate urban planning and urban biodiversity conservation, the habitat connectivity of the Township of Langley in British Columbia (BC), Canada will be analyzed in this research. The term “biodiversity” encompasses plants, animals, microorganisms, and the ecological processes of which they are parts (McNeely et al., 1990) and is generally defined as the species richness and evenness of a particular ecosystem. The importance of biodiversity has been addressed through its value in providing and sustaining actual and potential biological resources (WOOD, 1997). For instance, biodiversity is necessary for supporting ecosystem services, significant in improving human health and wealth (Sandifer et al., 2015), and critical in maintaining ecosystem stability and resilience at both spatial and temporal scales (Lohbeck et al., 2016). Urbanization, as one of the largest human-induced land use transformation processes, has been proven as a major threat to global and local biodiversity because it introduces long-lasting impacts to the ecosystem by cultivation, habitat fragmentation, deforestation, pollution, urban heat island effect, and so on (Garrard et al., 2018). In addition, due to the rapid development of human society, the expansion of urban areas is disproportionally faster than other types of land cover with forecasts for the future anticipating global growth of 200% by 2030 to support the surging urban human population (Elmqvist et al., 2013). Therefore, being the cost of land use transformation, the reduction in the city's natural environment and landscape is detrimental to the conservation and development of biodiversity on a global scale and urban scale.

Although natural environments support the majority of biodiversity on a global scale, the biodiversity of urban environments has also been addressed and studied in recent years because it is where urbanization takes place, and it is proven to support a large number of threatened species (Ives et al., 2016) and are critical space for biodiversity conservation. In the context of urban biodiversity, “habitat connectivity”, which refers to the degree to of habitat patches of individual species are functionally connected (Correa Ayram et al., 2016), has been addressed widely as one of the major components that sustain urban biodiversity. According to Crooks and Sanjayan (2006), habitat connectivity is essential in supporting the movement of species, materials, and energy across landscapes to preserve biodiversity and ecosystem services. Urban greenways such as parks, urban forests, and protected areas, are one crucial approach to maintaining natural connections and enhancing habitat connectivity, which is necessary for the preservation of larger breeding populations of plants and animals, more endangered species, more robust food webs, and more stable ecosystems. 

This research focuses on the quantification of habitat connectivity in the Township of Langley, BC, Canada, because the township is urbanized and fragmented, and habitat connectivity analysis is needed for urban planning. Habitat connectivity can be quantified as the probability of connectivity index (PC) (Saura & Pascual-Hortal, 2007), and this measure has been successfully used in different disciplines across the world at different scales. For instance, in Lanzhou, China, PC was used to assess the potential of green infrastructure networks in solving urban issues (Zhang et al., 2022), and for protected areas in China, PC was widely used in connectivity evaluation (Zhao et al., 2022). Despite the rising interest in biodiversity evaluation in the urban environment, there is still a lack of knowledge at the local scale in British Columbia (BC), Canada regarding the quantification of habitat connectivity. Furthermore, the Township of Langley locates within the Coastal Douglas-fir and Associated Ecosystem Conservation Partnership (CDFCP) boundary, and the municipality has a combination of two ecological zones, the Coastal Douglas-fir (CDF) Biogeoclimatic Classification Ecosystem (BEC) zone, and the Western Hemlock (CWH) BEC Zone. Both BEC zones are rare and valuable to Canada because they contain a significant number of endangered species and have high biodiversity. Research on habitat connectivity has been conducted across the world on different topics like freshwater ecosystems (Mahmoudzadeh et al., 2022), tiger conservation (Schoen et al., 2022), and ground-dwelling animals’ habitat (Braaker et al., 2014). Yet, habitat connectivity on CDF and CWH BEC zones in the context of urban areas has not been well understood. Therefore, the rare and valuable ecosystem, vulnerability, continuous urbanization, and active interaction between humans and the nature of the Township of Langley make it a suitable study area to conduct habitat connectivity quantification to facilitate urban planning. Urban planning and design advancements can lessen the effects of urbanization on urban biodiversity (Ibáñez-Álamo et al., 2020), and quantified measure of habitat connectivity is an urgent need for planners and developers to make sure that biodiversity conservation is considered when making decisions.

This study aims to analyze the functional habitat connectivity of three species with varying habitat requirements and mobility, utilizing the connectivity indexes derived from network analysis. The targeted species are the Western red-backed salamander (Plethodon vehiculum), the American red squirrel (Tamiasciurus hudsonicus), and the brown creeper (Certhia americana). In comparison with previous research conducted by William (2019) in the City of Courtenay which use similar approach, the hypothesized result of habitat connectivity is expected to the be similar be slightly lower in the Township of Langley due to the higher proportion of urban area and agricultural land. Three key objectives need to be met in this habitat connectivity analysis:

1. Analysis and comparison of the habitat connectivity of the three species.
2. Identification of important habitat patches in the Township of Langley.
3. Identification of the main ideal corridor for the Township of Langley.

Bordered by the Rocky Mountains on the east and the Pacific Ocean on the west, British Columbia is a coastal province located in western Canada that covers 944,735 square kilometers of land and freshwater ecosystems (Statistic Canada, 2022). Influenced by both the Pacific Ocean and the mountain ranges, BC's climate varies from region to region. The study area of this study (Figure 1), the Township of Langley, located at 49.0743° N, 122.5593° W, has a temperate marine climate, and a warm and humid climate in coastal areas (Welcome BC, 2022.). The Township of Langley is not directly adjacent to the ocean, resulting in hotter summers and colder winters compared to nearby coastal municipalities such as Vancouver and Richmond. The average elevation of the study area is 47m (topographic-map, 2022), the majority of which is built above Fraser's floodplain, leading most of the area susceptible to flooding.

Given its unique climate and topographical features, the Township of Langley is characterized by the combination of CDF and CWH BEC zones. The CDF zone is the smallest and most endangered ecological zone in Canada because it has the highest diversity of plants and overwintering birds (Ward et al., 1998) and has more species at risk than other ecological zones in British Columbia (Austin, 2008). Similarly, the CWH zone is also unique and rare in Canada. Due to its great diversity and abundance of habitat elements, the Fraser Lowland region of this zone contains the greatest variety of birds, amphibians, and reptiles in British Columbia, and the CWH contains nearly all the coastal habitats for colony-nesting birds (Pojar et al., 2022). Many species at risk are found inhabiting the CWH too, such as Vancouver Island Marmot, Northern Goshawk, and Spotted Owl (Pojar et al., 2022).

In addition to the unique geographical features, the anthropogenic features of the township are also worthy of attention. Although occupying only 9.5% of the area of Canada, BC have a population of 5,286,528 in 2021, accounting for 13.6 % of the total population of Canada (Statistic Canada, 2021). BC is populated, especially in the southwest of the province, the township accounts for 2.7% of the population (Township of Langley, 2022), and is one of the richest agricultural areas in Canada, with 75% of the land in the township are Agricultural Land Reserve (ALR) (Township of Langley, 2022). Being the primary practice in the township, agriculture not only sustains the local economy, and BC’s dairy product production, but also heavily influences the ecosystem structure, biodiversity, and climate of the region.

A raster dataset “LCC2014_5m_Hybrid” was obtained from the Metro Vancouver open data portal ( https://gisportal.metrovancouver.org/portal/apps/sites/#/open-data-portal ). This is a comprehensive land cover classification that provides a contiguous surface mapped to broad biophysical classes for Metro Vancouver and Abbotsford, with 5-meter spatial resolution. This classification was created using RapidEye 5m multi-spectral satellite imagery and full feature LiDAR data, wherever available. The imagery used in this classification dataset was from August 2014. The landcover classification is then reclassified into 6 classes (bare and soil, coniferous, deciduous, non-forest vegetation, urban, and water) for habitat selection purposes.

Light detection and ranging (LiDAR) is a remote sensing method in recording structural attributes including vegetation height, density and so on. Township of Langley has full coverage of light lidar data and is available in and free to access from the BC Lidar portal (LidarBC, 2021). The Township of Langley contains a total of 153 tiles of data, which were collected from June to September in 2016. An Optech Galaxy sensor was used to collect data from 1850m flying height (AGL) with a maximum scan angle of 30°. The resulting point data have an average point density of 7 points per square meter with an average non-vegetated vertical accuracy of 0.078m. A digital elevation model (DEM), also called digital terrain model (DTM) is generated from this dataset in R studio through the lidR package (Tompalski, 2022) using triangular irregular network algorithm and is rasterized at a spatial resolution of 5 m. The topographic wetness index as a raster layer is then generated from the DEM to determine the probability of moisture in the landscape. A canopy height model (CHM) with 5m resolution was also generated from the LiDAR dataset in R studio through the lidR package (Tompalski, 2022).

The Western red-backed salamander (Plethodon vehiculum), the American red squirrel (Tamiasciurus hudsonicus), and the brown creeper (Certhia americana) were selected as targeted species for a habitat connectivity study due to several reasons. Firstly, all three species are considered focal species, which are ecologically important and have a significant impact on their respective ecosystem (Diamond Head Consulting, 2022; Williams, 2019). Secondly, their habitats are threatened by human activities to different degrees (Berteaux & Boutin, 2000; Geleynse et al., 2016; Howard, 2003), making them ideal candidates for understanding the impact of urbanization on landscape habitat connectivity. Furthermore, these three species belong to different animal classes (amphibian, mammal, and bird) with different habitat requirements and dispersal ability. This choice is representative and beneficial for future ecology and biodiversity management because it allows us to study the connectivity needs of a range of taxa. For example, amphibians like the Western red-backed salamander are known to have limited dispersal ability, while birds like the brown creeper can disperse over larger distances with no ground feature limitation. This information can be used to design habitat corridors and prioritize conservation efforts in areas that will benefit multiple species. Overall, by studying the habitat connectivity of these three species, decision makers and city planners can identify key areas of importance for these species and develop strategies to manage and conserve their habitats.

Freshwater atlas. The wetland data of BC are obtained from the BC data catalog ( https://www2.gov.bc.ca/gov/content/data/geographic-data-services/topographic-data/freshwater  ) and reprojected projected in NAD 1983 UTM Zone 10N. This data was collected in the fall of 2009.

Road. The public road system for Township of Langley is obtained as a feature layer from the Township of Langley Open Data Portal ( https://data-tol.opendata.arcgis.com/ ) (2022). It is updated weekly and contains 3658 records showing information about road names, road length and so on.

Park. The public parks and greenways (regional, municipal, school and specialty) for Township of Langley are obtained as a polygon feature layer from the Township of Langley Open Data Portal ( https://data-tol.opendata.arcgis.com/ ) (2022).

Lidar data collected from the BC Lidar portal (LidarBC, 2021) was used to generate the digital elevation model (DEM) and the canopy height model (CHM) because it is essential to attribute height information to the trees, which relates to one of the habitat requirements of the creeper and the squirrel. To match the land cover layer’s resolution, both the CHM and DEM have a spatial resolution of 5 meters. To assess the probability of an area receiving moisture, the topographic wetness index (TWI) raster image was generated from the lidar-driven DEM model based on slope and height attributes on the landscape. The TWI provides information about how prone an area is to water accumulation based on the slope, aspect, and topography of the area (Kopecký et al., 2021). It was used to differentiate the moist area and the dry areas in this study. By querying the TWI value, a binary (1/0) image with the mean as the threshold is created. Pixel values in the binary image equal to 1 indicate moisture-receiving areas. This is useful information for habitat selection for long-toted salamanders since they inhabit moist areas.

The lidar-driven CHM was used to attribute height information to all forest classes (coniferous and deciduous forest) to estimate the extent of mature forest. According to William (2019), a height threshold of 24m was used to identify mature forests, which is favored by brown creepers and American red squirrels. Another binary image was created by querying the pixel value with this threshold, while pixel values equal to 1 indicate the mature forest.

To add more information for habitat selection, a 30m buffer zone was created around all roads. This approach corresponds to the unsuitability of habitat patches being too close to roads. Patches that are located within the 30 m road buffer zone will be filtered out.

Based on the patch characteristics mentioned above (moisture degree, tree maturity and proximity of road), with the requirements of habitat land cover class and the minimum patch size in Table 1, a theoretically suitable habitat patch layer for each species as generated.

Each habitat patch for the three species will be assigned a quality value. The quality value ranges from 1 to 3. It was assigned based on the proximity to wetlands and parks. A quality of 3 was assigned to patches that are within a 450 m buffer zone to wetlands and parks, a quality of 2 was assigned to patches that are within a 900 m buffer zone, and the rest will be assigned a quality of 1. The quality values of the entire landscape are shown in Figure 3. To create the area-weighted quality (a) value for each patch, a crucial input into Conefor for creating a connection network and modeling habitat connectivity, these quality values will be multiplied by the patch size.

The connection network between patches was established according to the maximum dispersion distance for each species (Table 1). Network creation was conducted in the Conefor plugin in ArcGIS (Jenness, 2016). Correspond to the dispersal limitation of long-toted salamanders and red squirrels identified in Table 1. Links that intersect with roads or highways will be removed manually. There is no link needed to be removed for the brown creeper because its dispersal is not limited by the presence of roads and highways. The output will be a connection file with the shortest distance that connected two patches based on the maximum dispersal distance of each species. The node files for each species were also generated by the Conefor plugin in ArcGIS, one habitat patch corresponds to one node.

The inputs for modeling the habitat connectivity are the nodes file and connection file, respectively for the 3 species. The overall habitat connectivity will be measured using the Probability of Connectivity (PC) index (Saura & Pascual-Hortal, 2007) in Conefor. PC is a graph-based habitat availability metric that allows the quantification of animals’ functional connectivity. The “probability that two points randomly placed within the landscape fall into habitat areas that are reachable from each other (interconnected) given a set of n habitat patches and the links (p _ij ) among them” (Saura & Pascual-Hortal, 2007):

To assess the relative significance of importance connectivity, nodes importance was generated from Conefor as well. They are the intra patch connectivity (dPCintra), patch connectedness (dPCflux), steppingstones importance (dPCconnector), effect of node removal on overall network connectivity (dPC) and patch importance in maintaining component integrity (dNC) (Saura & de la Fuente, 2017). dPC is the sum of dPCintra, dPCflux, and dPCconnector.

Important patches were identified into 3 categories based on each patch’s connectivity metrics, they are key patch, hub patch, and key-hub patch. Key patches are the most important patch in the overall connectivity for each species. The key patch is determined by dPC because it is a combination of within-patch and between-patch connections. The Hub patches are the most critical patches for each species in terms of sustaining network connections. It is determined by dPCconnector and dNC. The key-hub patches are patches that are important both in overall connectivity and in maintaining network connectivity. For all species, key patches were selected at the 80 ^th  percentile of dPC, while hub patches were selected at 90 ^th  percentile of dPCconnector and all patches with a negative dNC. Key-hub patches were the overlap of these two categories.

The main connectivity corridor and area for improvement of the Township of Langley were identified. Ideally and theoretically, corridors are routes that connect the most important habitats in the landscape, and where the highest success rate of dispersal could be found. The corridor was manually digitized based on the key-hub patches for all species. Shortest connection distance between patches and large patch size were considered in digitization. Areas of improvement were symbolized as pinch points; they are manually identified based on the intersection of highway. In addition, large connection gaps between key-hub patches, which were considered to greatly improved the network if connection presents, are also identified as pinch points. Failure of dispersal of some species might found at these pitch points.

Figure 6 shows the habitat patches for all three species symbolized by dPC value of each patch, which is calculated based on the area-weighted quality and the dispersal distance. The majority of patches have low dPC values, and a very limited number of patches have high dPC values; this same pattern is found over all three species. Generally, there is no observable spatial distribution pattern of the value of dPC of these patches. However, a patch located at the southwestern corner of the landscape is found to have a high dPC value for both the squirrel and the brown creeper. Similar pattern is found in the medium patches too. Similar findings were found in patches with medium dPC value in the same area. The dPC value distribution of the salamander is less in common with the squirrel and the brown creeper regarding to the medium and high dPC values.

Table 2 shows the number of patches that were identified as key patch, nub patch and key-hub patches. Figure 7 shows the key patches for three species, which are selected at the 80 ^th  percentile of all patches base on the dPC value, therefore with the combination of within-patch and between-patch connectivity in determining overall patch importance, key patches are the most important patches overall for each species. Hub patches for each species were identified by the 90 ^th  percentile of dPCconnector and negative dNC are shown in figure 8. Hub patches are the most important patches for each species for maintaining the connectivity of their respective networks, the removal of these patches will lead to an increase of habitat fragmentation. In terms of key patches, the distribution of the squirrel’s key patches is mainly located in the southwest corner, the central and the northern parts of the city. The patches in the southwest corner are more clustered together. The key patches for the brown creeper are all located at the southwest corner of the landscape. The key patches of the salamander, on the other hand, were relatively evenly distributed throughout the landscape. Similar distribution pattern of hub patches is found in the squirrel and salamander. Yet there are two brown creeper’s hub patches located at the northeast corner of the landscape.

Figure 9 shows the key-hub patches of the three species, which are the overlap patches in the “key” and “hub” categories. These patches build the foundation of the digitization of the landscape connectivity corridor. Similar with the spatial distribution pattern of key patch, the distribution of key-hub patches for the squirrel and the brown creeper is more clustered, while key-hub patches are found to be more evenly distributed across the space.

Figure 10 shows the common key-hub patches for the three species and the union key-hub patches of all species in light green. there are two patches that fall into the same “key-hub” categories for the three species at the same time. The squirrel and salamander share 161 key-hub patches, which is half of their patch total. Although there are only 3 overlapped key-hub patches between the brown creepers and the squirrel, all its key-hub patches overlap with those of the squirrel. The brown creeper has the highest key-hub patch overlapping rate as 67% of the brown creeper’s key-hub patches overlapped with which of other species. The common key-hub patches for all species are located at the southwestern corner of the landscape.

With increasing urbanization, habitat fragmentation has become a common issue, leading to the loss of biodiversity. This study addresses this problem by analyzing the functional habitat connectivity of the American squirrel, brown creeper, and Western red-backed salamander in the Township of Langley, BC. In this study, the overall habitat connectivity of the three species in the study area was examined through a network analysis with the probability of connectivity index (PC). In addition, based on the patch importance indexes, key patches, hub patches, and key-hub patches for each species were identified and mapped to facilitate patch-level management purposes. Finally, a connectivity corridor across the corridor was highlighted, this corridor connected the most critical patches for all species in the landscape, aiming to provide a sense of knowledge to city planners about the potential dispersal route of the species and location for future greenway construction.

The key finding of this research is that the habitat connectivity of the animal species with different proxies, including the American red squirrel, the brown creeper, and the red-backed salamander, is low in the Township of Langley, and the difference between them are significant. The result suggests that habitat connectivity of amphibians, small mammals and birds are low in urban areas due to the limitation of habitat requirement, dispersal distance and dispersal restriction. Despite having the lowest maximum dispersal distance among the studied species, the salamander exhibits the highest probability of connectivity, which can be attributed to the extensive coverage of its habitat patches, resulting in the highest chance to fall into areas that are reachable from each other. Despite traveling the furthest distance, the creeper exhibits the lowest habitat connectivity due to its high requirement in habitat selection. Specifically, the creeper 's preference for certain habitats based on tree size and maturity resulted in a limited number of suitable patches in the landscape, with only six such patches identified in total.

The result of this analysis is then compared with the landscape connectivity studies (Diamond Head Consulting, 2022; Williams, 2019) incorporating the same method and same species, the result of red squirrel, the brown creeper is found to be similar at scale but significantly lower at value. The observed differences in habitat connectivity may be attributed to the nature of the landscape studied. The township of Langley is characterized by high levels of urbanization and significant agricultural land use, resulting in fragmented potential habitat patches. These factors may have limited the movement of organisms between patches, resulting in a relatively lower habitat connectivity values compared to studies conducted in more natural or less fragmented landscapes. In addition, it is speculated that this might be due to the difference in quality value assignment to habitat patches and the variation when incorporating disposal limitation of species. In this study, highways were considered as a potential dispersal limitation for red squirrels, which may have resulted in lower connectivity values compared to other studies that did not consider this factor. One unanticipated finding was that the habitat connectivity of the red-backed salamander is substantially lower than when the analysis is scaled to the entire Metro Vancouver. One possible explanation for this difference is that a different set of criteria for selecting potential habitat patches based on landcover types was used in this study. All other parameters used in this analysis were consistent with those used in other studies, suggesting that the approach of selecting habitat patches for the red-backed salamander played a key role in the divergent results. Overall, the findings suggest that habitat fragmentation resulting from high levels of urbanization and significant agricultural land use has a significant impact on the movement of organisms between patches, resulting in relatively lower habitat connectivity values compared to studies conducted in more natural or less fragmented landscapes. This result has important implications for urban planning and conservation efforts, highlighting the need to consider the effects of human activities on wildlife habitat connectivity and to develop strategies to mitigate the negative impacts of urbanization on wildlife populations.

To gain a deeper understanding of the habitat connectivity of each species the summary of different patch importance index, a product of network analysis, was investigated and mapped. As shown in the result, apart from the variations of species in overall habitat connectivity, measured by the probability of connectivity index (PC), the relative importance of specific patches in promoting connectivity varies among species as well. Among the studied species, the squirrel exhibits the highest sumdPC, indicating that its network contains a greater number of patches whose removal would significantly decrease network connectivity by causing parts of the network to become disconnected. However, the salamander exhibits the smallest sumdPC, despite having the largest total probability of connectivity, due to its lowest dispersal distance, which means that the removal of some patches might not significantly affect the overall network if they are not connected to other patches. In general, this this result shows that the impact of removing patches for different species is an important consideration in urban planning and ecology management. Hence, when planning and managing urban landscapes, city planners and ecologists should consider the specific needs and requirements of different species, rather than applying a one-size-fits-all approach.

Figure 5 reveals that most of the American red squirrel's sumdPC is contributed by sumdPCintra, indicating habitat patches for the squirrel have a very high contribution to within-patch connectivity. Furthermore, the squirrel shows the highest sum dPCconnector value, suggesting that steppingstones play a greater role in network connectivity than which of the other two species. Meanwhile, the brown creeper exhibits the highest sumdPCflux, suggesting that its patches are highly connected despite their small number. This is likely due to the species' longest dispersal distance and the absence of dispersal limitations. Surprisingly, the squirrel has the lowest sumdPCflux, suggesting that individual patches are not as connected as the other species to the rest of the network. One of the possible explanations to this result is that although it has the second longest dispersal distance, there are many patches, so even many patches are connected, it is relatively low when considering the entire picture. Additionally, because the dispersal is restricted by the highway, the connectivity of one patch to the rest of the network is further reduced. Overall, to promote connectivity, creating and preserving interconnected patches, steppingstones, and corridors between patches could play an important role. Additionally, these findings highlight the importance of understanding the species-specific factors that affect patch importance in promoting habitat connectivity, such as dispersal distance, habitat patch size and distribution, and the presence of dispersal barriers. For future development of the city, locations of road and highway construction should be considered carefully in reference to wildlife’s dispersal limitation.

The major deliverable of this study is to generate a theoretical connectivity corridor to the landscape in which the dispersal success is estimated to be the highest based on the most critical patches for all species, the corridor is mapped in Figure 11. The corridor spans the entire landscape from south to north and from east to west. However, these connected habitat patches are mainly contributed by the salamander’s, and only the southwest of the landscape would be depicted if the common key-hub habitat patches for all species were considered (Figure 10). Therefore, this connectivity corridor is built on the key-hub patches of all species, rather than their common patches. The areas requiring connectivity enhancement are also identified as pinch points in Figure 11. The highway pinch points include pinch points located on Trans-Canada Highway number 1 and number 13. Improving habitat connectivity by building green corridors across the highway allows safer and easier movement between habitat patches and connects the northern part to the rest of the township. Improving the connectivity in the Belmont golf course pinch point can connect the northwester clusters to the major corridor. However, there are forests and vegetation presented in that pinch point, the possible reason why it was not connected is that the forest might not be mature, so it falls outside of the scope of habitat requirement for both the squirrel and creeper. The pinch point that located at Trinity Western University can be greatly improved as it is very urbanized. Connected green corridors like landscaping trees might be one of the approaches to enhance connectivity. The Highline Mushroom Farm pinch point is another area that needs attention for connectivity improvement. In conclusion, this study highlights the importance of understanding species-specific factors that affect patch importance in promoting habitat connectivity. The identified pinch points in the landscape, particularly those located on highways, golf courses, and urbanized areas, provide opportunities for city planners to improve habitat connectivity by building green corridors and enhancing existing vegetation. Furthermore, urban planners can also put effort into improving the quality of habitat patches to increase their attractiveness to species and potentially improve the likelihood of successful dispersal. These efforts can help improve habitat connectivity, support the viability and resilience of wildlife populations in urban areas, and promote long-term ecological management of the landscape and conservation of biodiversity.

This study was subject to several limitations that could have impacted on the results. Firstly, due to limited time and resources, an outdated landcover dataset from 2014 was utilized during habitat patch selection. The 9-year gap between the landcover data used in this study and the present day may have led to inaccuracies in results, particularly if there have been major changes to the township since 2014. While it is difficult to predict the magnitude and direction of these potential discrepancies, conducting a land cover classification with more recent data would help to address this limitation.

Secondly, the high number of patches and links required for connectivity analyses proved to be computationally expensive, with some species analyses taking up to 3 days to run. As a result, this study was forced to modify species parameters for the squirrel, the minimum patch size was increased from 500 to 900 m ² , which could have led to an underestimation of habitat connectivity. Future studies should consider using more powerful computing resources or alternative software to mitigate this limitation.

Lastly, the study did not fully account for dispersal barriers such as buildings or topographic features when incorporating dispersal distances. This limitation could have resulted in an overestimation of connectivity between patches. To improve future connectivity studies, it is recommended to identify major barriers and incorporate road systems when generating links between patches. Overall, while these limitations may have impacted the accuracy of the results, it is believed that this study provides valuable insights into habitat connectivity for the Township of Langley.

In summary, the study on habitat connectivity and ecological management in Langley Township was limited by the use of an outdated land cover data set, the computational power of network analysis, and the limited inclusion of barriers to dispersal. These limitations may have affected the accuracy of the results, leading to the possibility that habitat connectivity may be underestimated or overestimated. City planners and ecologists should be aware of the potential inaccuracies of outcomes when making decisions to improve habitat connectivity. Using more up-to-date land cover data, more powerful computing resources, and incorporating additional potential dispersal barriers are useful measures to produce more accurate results. Despite these limitations, this study provides valuable insights into habitat connectivity in Langley Township and serves as a basis for future studies of habitat connectivity, ecological management, and biodiversity conservation in urban environments.

B.C. Ministry of Environment. (2015). Management Plan for the Northern Red-legged Frog (Rana aurora) in British Columbia. B.C. Ministry of Environment.  https://a100.gov.bc.ca/pub/eirs/finishDownloadDocument.do?subdocumentId=10251 

Berteaux, D., & Boutin, S. (2000). Breeding Dispersal in Female North American Red Squirrels. Ecology, 81(5), 1311–1326.  https://doi.org/10.1890/0012-9658(2000)081[1311:BDIFNA]2.0.CO;2 

Biogeoclimatic Zones of British Columbia. (n.d.). Retrieved October 1, 2022, from  https://www.for.gov.bc.ca/hfd/library/documents/treebook/biogeo/biogeo.htm 

Braaker, S., Moretti, M., Boesch, R., Ghazoul, J., Obrist, M. K., & Bontadina, F. (2014). Assessing habitat connectivity for ground-dwelling animals in an urban environment. Ecological Applications, 24(7), 1583–1595.  https://doi.org/10.1890/13-1088.1 

Bunnell, F., Chan-McLeod, A., Sutherland, G., & I, H. (2000). Persistence of forest-dwelling vertebrates in managed forests.  https://doi.org/10.13140/RG.2.2.19428.68486 

Cai, G., Ren, H., Yang, L., Zhang, N., Du, M., & Wu, C. (2019). Detailed Urban Land Use Land Cover Classification at the Metropolitan Scale Using a Three-Layer Classification Scheme. Sensors, 19(14), Article 14.  https://doi.org/10.3390/s19143120 

Chytyk, P., & Fraser, D. F. (2015). Brown Creeper. The Atlas of the Breeding Birds of British Columbia, 2008-201.  https://www.birdatlas.bc.ca/ 

Correa Ayram, C., Mendoza, M., Etter, A., & Pérez-Salicrup, D. (2016). Habitat connectivity in biodiversity conservation: A review of recent studies and applications. Progress in Physical Geography, 40, 7–37.  https://doi.org/10.1177/0309133315598713 

Crooks, K. R., & Sanjayan, M. (Eds.). (2006). Connectivity Conservation. Cambridge University Press.  https://doi.org/10.1017/CBO9780511754821 

Dumitru, C. O., Schwarz, G., Pulak-Siwiec, A., Kulawik, B., Albughdadi, M., Lorenzo, J., & Datcu, M. (2020). Understanding satellite images: A data mining module for Sentinel images. Big Earth Data, 4(4), 367–408.  https://doi.org/10.1080/20964471.2020.1820168 

Elmqvist, T., Fragkias, M., Goodness, J., Güneralp, B., Marcotullio, P. J., McDonald, R. I., Parnell, S., Schewenius, M., Sendstad, M., Seto, K. C., & Wilkinson, C. (Eds.). (2013). Urbanization, Biodiversity and Ecosystem Services: Challenges and Opportunities. Springer Netherlands.  https://doi.org/10.1007/978-94-007-7088-1 

Garrard, G. E., Williams, N. S. G., Mata, L., Thomas, J., & Bekessy, S. A. (2018). Biodiversity Sensitive Urban Design. Conservation Letters, 11(2), e12411.  https://doi.org/10.1111/conl.12411 

Geleynse, D. M., Nol, E., Burke, D. M., & Elliott, K. A. (2016). Brown Creeper (Certhia americana) demographic response to hardwood forests managed under the selection system. Canadian Journal of Forest Research, 46(4), 499–507.  https://doi.org/10.1139/cjfr-2015-0112 

Ibáñez-Álamo, J. D., Morelli, F., Benedetti, Y., Rubio, E., Jokimäki, J., Pérez-Contreras, T., Sprau, P., Suhonen, J., Tryjanowski, P., Kaisanlahti-Jokimäki, M.-L., Møller, A. P., & Díaz, M. (2020). Biodiversity within the city: Effects of land sharing and land sparing urban development on avian diversity. Science of The Total Environment, 707, 135477.  https://doi.org/10.1016/j.scitotenv.2019.135477 

Ives, C. D., Lentini, P. E., Threlfall, C. G., Ikin, K., Shanahan, D. F., Garrard, G. E., Bekessy, S. A., Fuller, R. A., Mumaw, L., Rayner, L., Rowe, R., Valentine, L. E., & Kendal, D. (2016). Cities are hotspots for threatened species. Global Ecology and Biogeography, 25(1), 117–126.  https://doi.org/10.1111/geb.12404 

Jenness, J. (2016, April 9). Jenness Enterprises—ArcGIS Tools; Conefor Inputs Tool.  http://www.jennessent.com/arcgis/conefor_inputs.htm 

Johnson, B. A. (2013). High-resolution urban land-cover classification using a competitive multi-scale object-based approach. Remote Sensing Letters, 4(2), 131–140.  https://doi.org/10.1080/2150704X.2012.705440 

Kamusoko, C. (2022). Land Cover Classification Accuracy Assessment. In C. Kamusoko (Ed.), Optical and SAR Remote Sensing of Urban Areas: A Practical Guide (pp. 105–118). Springer.  https://doi.org/10.1007/978-981-16-5149-6_6 

Larsen, K. W., & Boutin, S. (1994). Movements, Survival, and Settlement of Red Squirrel (Tamiasciurus Hudsonicus) Offspring. Ecology, 75(1), 214–223.  https://doi.org/10.2307/1939395 

Leutner, B., Benjamin, N., Schwalb-Willman, J., & Hijmans, R. J. (2022, March 7). RStoolbox: Tools for Remote Sensing Data Analysis in R.  https://bleutner.github.io/RStoolbox/ 

LidarBC. (2021, November). LidarBC - Open LiDAR Data Portal.  https://governmentofbc.maps.arcgis.com/apps/MapSeries/index.html?appid=d06b37979b0c4709b7fcf2a1ed458e03 

LidarBC. (2021, November). LidarBC - Open LiDAR Data Portal.  https://governmentofbc.maps.arcgis.com/apps/MapSeries/index.html?appid=d06b37979b0c4709b7fcf2a1ed458e03 

Lohbeck, M., Bongers, F., Martinez-Ramos, M., & Poorter, L. (2016). The importance of biodiversity and dominance for multiple ecosystem functions in a human-modified tropical landscape. Ecology, 97(10), 2772–2779.  https://doi.org/10.1002/ecy.1499 

Mahmoudzadeh, H., Masoudi, H., Jafari, F., Khorshiddoost, A. M., Abedini, A., & Mosavi, A. (2022). Ecological networks and corridors development in urban areas: An example of Tabriz, Iran. Frontiers in Environmental Science, 10, 969266.  https://doi.org/10.3389/fenvs.2022.969266 

Mahmoudzadeh, H., Masoudi, H., Jafari, F., Khorshiddoost, A. M., Abedini, A., & Mosavi, A. (2022). Ecological networks and corridors development in urban areas: An example of Tabriz, Iran. Frontiers in Environmental Science, 10, 969266.  https://doi.org/10.3389/fenvs.2022.969266 

Maxcy, K. A. (2004). RED-LEGGED FROG Rana aurora aurora. Accounts and Measures for Managing Identified Wildlife.  https://www.env.gov.bc.ca/wld/frpa/iwms/documents/Amphibians/a_redleggedfrog.pdf  \

McNeely, J. A., Miller, K., Mittermeier, R. A., Reid, W. V., & Werner, T. B. (1990). Conserving the world’s biological diversity. IUCN.  https://portals.iucn.org/library/node/5888 

Pauli, J. N. (2005). Evidence for Long-Distance Swimming Capabilities in Red Squirrels, Tamiasciurus hudsonicus. Northeastern Naturalist, 12(2), 245–248.

Pojar, J., Klinka, K., & Demarchi, D. A. (n.d.). Chapter 6: Coastal Western Hemlock Zone. 17.

Reference

Sandifer, P. A., Sutton-Grier, A. E., & Ward, B. P. (2015). Exploring connections among nature, biodiversity, ecosystem services, and human health and well-being: Opportunities to enhance health and biodiversity conservation. Ecosystem Services, 12, 1–15.  https://doi.org/10.1016/j.ecoser.2014.12.007 

Saura, S., & de la Fuente, B. (2017). Connectivity as the Amount of Reachable Habitat: Conservation Priorities and the Roles of Habitat Patches in Landscape Networks. In S. E. Gergel & M. G. Turner (Eds.), Learning Landscape Ecology: A Practical Guide to Concepts and Techniques (pp. 229–254). Springer.  https://doi.org/10.1007/978-1-4939-6374-4_14 

Saura, S., & Pascual-Hortal, L. (2007). A new habitat availability index to integrate connectivity in landscape conservation planning: Comparison with existing indices and application to a case study. Landscape and Urban Planning, 83(2–3), 91–103. https://doi.org/10.1016/j.landurbplan.2007.03.005

Saura, S., & Pascual-Hortal, L. (2007). A new habitat availability index to integrate connectivity in landscape conservation planning: Comparison with existing indices and application to a case study. Landscape and Urban Planning, 83(2–3), 91–103.  https://doi.org/10.1016/j.landurbplan.2007.03.005 

Sørensen, R., Zinko, U., & Seibert, J. (2006). On the calculation of the topographic wetness index: Evaluation of different methods based on field observations. Hydrology and Earth System Sciences Discussions, 10(1), 101–112.

Statistic Canada. (n.d.). Table 15.7 Land and freshwater area, by province and territory. Retrieved October 29, 2022, from  https://www150.statcan.gc.ca/n1/pub/11-402-x/2010000/chap/geo/tbl/tbl07-eng.htm#archived 

Steele, M. A. (1998). Tamiasciurus hudsonicus. Mammalian Species, 586, 1–9.  https://doi.org/10.2307/3504443 

Stefanov, W. L., Ramsey, M. S., & Christensen, P. R. (2001). Monitoring urban land cover change: An expert system approach to land cover classification of semiarid to arid urban centers. Remote Sensing of Environment, 77(2), 173–185.  https://doi.org/10.1016/S0034-4257(01)00204-8 

The European Space Agency. (2022). Sentinel-2—Missions—Resolution and Swath—Sentinel Handbook—Sentinel Online.  https://sentinels.copernicus.eu/web/sentinel/missions/sentinel-2/instrument-payload/resolution-and-swath 

Tompalski, J.-R. R., Tristan R. H. Goodbody, Piotr. (2022a). 4 Digital terrain model | The lidR package.  https://r-lidar.github.io/lidRbook/dtm.html#dtm 

Tompalski, J.-R. R., Tristan R. H. Goodbody, Piotr. (2022a). 4 Digital terrain model | The lidR package.  https://r-lidar.github.io/lidRbook/dtm.html#dtm 

Tompalski, J.-R. R., Tristan R. H. Goodbody, Piotr. (2022b). 5 Height normalization | The lidR package.  https://r-lidar.github.io/lidRbook/norm.html 

Tompalski, J.-R. R., Tristan R. H. Goodbody, Piotr. (2022b). 5 Height normalization | The lidR package.  https://r-lidar.github.io/lidRbook/norm.html 

topographic-map. (n.d.). Township of Langley topographic map, elevation, terrain. Topographic Maps. Retrieved October 29, 2022, from  https://en-ca.topographic-map.com/map-j5gz4/Township-of-Langley/ 

Township of Langley Open Data Portal. (2022).  https://data-tol.opendata.arcgis.com/ 

Township of Langley. (2022). About the Township. Retrieved October 29, 2022, from  https://www.tol.ca/your-township/about-the-township/ 

Township of Langley. (2022). Agriculture in Langley. Retrieved October 29, 2022, from  https://www.tol.ca/your-township/about-the-township/agriculture-in-langley/ 

Ward, P. (1998). Sensitive ecosystems inventory, east Vancouver Island and Gulf Islands, 1993-97, vol. 1: Methodology, ecological descriptions and results. Environment Canada, Environmental Conservation Branch, Canadian Wildlife Service, Pacific & Yukon Region.

WelcomeBC. (n.d.). Climate of BC. Retrieved October 29, 2022, from  https://www.welcomebc.ca/Choose-B-C/Explore-British-Columbia/Climate-of-B-C 

Williams, D. (2019). City of Courtenay Landscape Connectivity Analysis. Diamond Head Consulting.

Williams, D. (2019). City of Courtenay Landscape Connectivity Analysis (p. 46). Diamond Head Consulting.

WOOD, P. M. (1997). Biodiversity as the Source of Biological Resources: A New Look at Biodiversity Values. Environmental Values, 6(3), 251–268.

Yan, W. Y., Shaker, A., & El-Ashmawy, N. (2015). Urban land cover classification using airborne LiDAR data: A review. Remote Sensing of Environment, 158, 295–310.  https://doi.org/10.1016/j.rse.2014.11.001 

Zhang, X., Ren, Y., Zhang, D., & Li, K. (2022). Construction of the green infrastructure network for adaption to the sustainable future urban sprawl: A case study of Lanzhou City, Gansu Province, China. Ecological Indicators, 145, 109715.  https://doi.org/10.1016/j.ecolind.2022.109715 

Zhao, Z., Wang, P., Wang, X., Wang, F., Tseng, T.-H., Cao, Y., Hou, S., Peng, J., & Yang, R. (2022). A Protected Area Connectivity Evaluation and Strategy Development Framework for Post-2020 Biodiversity Conservation. Land, 11(10). Scopus.  https://doi.org/10.3390/land11101670