Jiangsu Province, located in eastern China’s Yangtze River Delta, is bordered to the east by the Yellow Sea. Its coastal cities, from north to south, are Lianyungang, Yancheng, and Nantong. The region hosts three major ecosystems: forests, oceans, and wetlands, with unique sandy ridges and intertidal wetlands. The terrestrial study area is defined by the administrative boundaries of these coastal cities, while the marine area covers waters under Jiangsu’s jurisdiction (Figure 1).

The Habitat Quality module in the InVEST model was used to assess the habitat quality of the study area. The model is based on habitat types, establishing a connection between habitat quality and threat sources. It analyzes habitat sensitivity and the impact of external threat factors, thereby providing insights into the spatial distribution of habitat quality in the region. The habitat quality index, ranging from 0 to 1, is used to represent regional habitat quality, with higher values indicating better habitat quality. The model’s calculation principles are detailed in the user guide. The calculation formula for habitat quality is as follows:

Here, Q_xj represents the habitat quality of raster cell x within land (or marine) use type j.  Hj denotes the habitat suitability of land (or marine) use type j, reflecting the capacity of different habitat types to provide suitable conditions for species’ survival and reproduction. z is a normalization constant, set to 2.5, and K is the half-saturation constant, with a default value of 0.5. Dxj is the threat index of raster cell x within land (or marine) use type j, used to quantify the level of stress imposed by threat factors and the degree of habitat degradation (Nature Capital Project, 2025).

Land-use data for habitat classification were obtained from the Resources and Environmental Science Data Center, Chinese Academy of Sciences (http://www.resdc.cn). Marine resource development data were provided by the Jiangsu Department of Natural Resources. Benthic biodiversity data came from surveys and ecological monitoring (2006-2020), with station locations shown in Figure 2.

Terrestrial habitat quality factors included annual average temperature, soil type, and DEM data from the Resources and Environmental Science Data Center. Precipitation data were from the Earth Resources Data Cloud (http://www.gis5g.com). Population density data were from WorldPop (https://hub.worldpop.org/), and nighttime lighting data were from Wu et al. (2022). GDP data combined per capita GDP and population density.

Marine habitat quality factors included average surface current velocity from the Hybrid Coordinate Ocean Model (HYCOM), sediment type data from Mei et al. (2020), seafloor topography from the Jiangsu Department of Natural Resources, and seawater eutrophication index from field sampling, as shown in Figure 2. Shipping and fishing data were from the Global Maritime Traffic Density Service (GMTDS).

Figure 3A illustrates the spatial distribution pattern of habitat quality along the Jiangsu coastal zone, with habitat quality ranging from 0 to 1. It is classified into five levels: low [0, 0.2), lower [0.2, 0.4), moderate [0.4, 0.6), higher [0.6, 0.8), and high [0.8, 1]. Low-quality habitats were mainly distributed in the central urban areas of the three coastal cities and near-shore reclamation zones. Lower-quality habitats are mainly found in dryland and the surrounding paddy fields of urban centers. Moderate-quality habitats are primarily located in dryland and near-shore mudflats, while higher and high-quality habitats are mainly found in marine areas. Overall, habitat quality along the Jiangsu coast shows a gradual increase from land to sea, with lower and moderate-quality habitats predominantly on land, and higher and high-quality habitats predominantly in the marine areas. Note that in this study, the “coastal zone” refers to the entire spatial extent of the study area along the Jiangsu coast. The terrestrial and marine areas are separated by the officially coastline provided by government. Together, these two areas constitute the coastal zone.

From 2010 to 2020, habitat quality along the Jiangsu coastal zone showed a declining trend, with terrestrial habitat quality decreasing more gradually and marine habitat quality experiencing a more significant decline, particularly between 2010 and 2015 (Figure 3B). Kernel density estimation was used to further analyze the temporal dynamics of habitat quality from 2010 to 2020. The results indicated that habitat quality along the Jiangsu coast consistently exhibited multipolarization, with this feature becoming increasingly pronounced. The kernel density curve for terrestrial habitats showed a stable three-peak pattern from 2010 to 2020, with stable distances between peaks, while the height of peak P1 gradually increased. Additionally, the right tail of the density curve gradually diminished. For marine habitats, the kernel density curve was unimodal in 2010 but transitioned to a bimodal distribution in 2015 and 2020. The main peak (P6) decreased in height, and a new peak (P4) emerged at 0.6, with P6 continuing to decline while P4 steadily increased (Figure 3C).

Between 2010 and 2015, habitat degradation zones were mainly concentrated in the central marine areas of Jiangsu, accounting for 16.45% of the total study area. From 2015 to 2020, this proportion decreased to 13.19%, with degradation zones primarily concentrated in the northern marine areas of Yancheng and the southern marine areas of Nantong. The overall analysis from 2010 to 2020 indicates significant habitat degradation in the central and southern marine areas, which accounted for 19.12% of the total study area. Overall, habitat quality degradation along the Jiangsu coast displayed notable spatial heterogeneity, with varying trends across different time periods. Degraded habitats were mainly concentrated in marine areas, while terrestrial degradation zones were relatively dispersed (Figure 3D).

The results of this study revealed that between 2010 and 2020, habitat quality in the Jiangsu coastal zone showed a clear degradation trend, characterized by distinct land-sea differences in the degradation process. On land, the expansion of low-quality habitats led to the encroachment of natural habitats, while in the marine environment, high-quality habitat areas transitioned toward regions with lower habitat quality.

The expansion of artificial habitats on land, which encroached upon natural areas, was a key indicator of habitat quality degradation in the coastal zone. Urbanization and the resulting encroachment on natural habitats are the primary drivers of this degradation (Wu et al., 2021a). Urban expansion can directly or indirectly lead to the loss of natural habitats (van Vliet, 2019). In the study area, urban expansion primarily contributed to habitat loss indirectly. The terrestrial habitat quality degradation zones in the Jiangsu coastal zone are mainly located in the agricultural areas surrounding urban development zones, although their spatial extent is relatively small. The coastal areas of Jiangsu have a high level of urbanization, and the demand for the expansion of built-up land has decreased. Additionally, the optimization and upgrading of the industrial structure have played a positive role in this process. As traditional manufacturing industries gradually transition to high-tech sectors, the demand for land-intensive industrial expansion has significantly decreased. Moreover, the development of the tertiary sector increasingly relies on existing urban infrastructure rather than large-scale new land development. Research has shown that urbanization in the coastal areas of Jiangsu is highly coordinated with habitat quality (Li et al., 2024). Furthermore, the rate of terrestrial habitat quality degradation in the Jiangsu coastal zone has slowed since 2015. Since 2014, China’s land-use model has gradually shifted from extensive to more intensive and efficient patterns (Liu et al., 2018). This transition has, to some extent, curbed the rapid expansion of built-up land, slowing habitat quality degradation and promoting the efficient use of land resources.

The kernel density map of habitat quality in the study area’s marine environment revealed a distinct unimodal pattern in 2010, which gradually transformed into a bimodal distribution by 2015 and 2020, with a new peak emerging at 0.6. The height of this new peak continuously increased, while the height of the original peak at 0.8 progressively decreased. This shift suggested that the transition of high-quality marine habitat areas toward relatively higher-quality zones represented another manifestation of habitat quality degradation in the coastal zone. The spatial distribution map of habitat quality degradation indicated a strong correlation between marine habitat degradation zones and areas of declining benthic biodiversity, while there was little relationship with regions experiencing significant changes in development activity intensity. Furthermore, the spatial distribution of these degradation zones exhibited considerable irregularity, reflecting the indirect impact mechanism of marine development activities on habitat quality degradation. Benthic organisms, as sensitive ecological indicators, were capable of directly reflecting changes in water quality, sediment characteristics, and overall ecosystem stability (Jayachandran et al., 2020). Although development activities such as aquaculture and offshore wind farms had limited direct disturbance to localized areas, they could have profound effects on marine habitat quality by altering the overall marine environmental conditions, including water quality and sediment characteristics (Farr et al., 2021; Hao et al., 2024). The three-dimensional dynamics of marine space further exacerbated the irregular spatial distribution of these degradation zones.

Overall, the expansion of low-quality habitat areas on land, which encroach upon natural habitats, and the transition of high-quality marine habitat areas toward relatively lower-quality zones, both contributed to habitat quality degradation in the coastal zone.

Climate change and human activities are two primary threats driving global biodiversity loss, with habitat quality responding differently to human activities and natural environmental changes across regions (Calderon-Aguilera et al., 2012; Phillips et al., 2017). The results of this study indicate that habitat quality in the coastal zone exhibits land-sea differences in its response to human activities and natural environmental changes: terrestrial habitats are primarily driven by human activities, whereas marine habitats are mainly governed by natural factors, and exhibit nonlinear responses under the influence of human activities.

From 2010 to 2020, terrestrial habitat quality exhibited a compound driving pattern characterized by human dominance with natural factors playing a supporting role. Results from the optimal-parameter Geodetector analysis indicated that nighttime light intensity, a typical proxy for human activity, was the primary driving factor across all three periods, with its explanatory power showing an ‘increasing–high-level stabilization’ trend. Specifically, in 2010, both temperature and nighttime light had relatively high explanatory power, followed by soil type, indicating a notable contribution of natural factors. By 2015, nighttime light and population density became the dominant factors, while the explanatory power of natural factors declined, reflecting a rapid intensification of human disturbances. In 2020, the explanatory power of nighttime light further increased, with GDP and population density following, establishing a stable, human-dominated driving pattern. Studies have shown that human activities are the primary driver of land habitat quality changes (Luo et al., 2024; Lin et al., 2024; Shi et al., 2024), primarily due to their long-term, widespread, and profound interventions in terrestrial ecosystems (Knapp et al., 2017). Regarding interaction mechanisms, human activities and natural factors exhibited a pronounced nonlinear enhancement effect. Although natural variables showed relatively low explanatory power in the single-factor analysis, they still played an important supporting role when interacting with indicators of human disturbance such as nighttime light intensity and population density. Due to inherent environmental conditions and long-standing agricultural practices, cropland has remained the dominant land type in the terrestrial area of the study region. While the intensification of economic development and construction activities continues to exert direct pressure on habitat quality, the strong sensitivity of agricultural land to natural factors—such as temperature, precipitation, and soil salinity—ensures that climatic factors still exert a critical influence on terrestrial habitat dynamics (Corwin, 2021). Although natural factors had weaker explanatory power than human activities in the single-factor analysis, they demonstrated a markedly enhanced effect in interactive combinations.

The driving mechanisms of marine habitat quality exhibit a phased pattern characterized as ‘natural-factor dominance → intensified human disturbance → return to natural-factor dominance.’ Single-factor analyses show that in 2010, marine habitat quality was primarily driven by DEM and marine development activities, indicating strong constraints from geomorphological settings alongside localized anthropogenic disturbances. In 2015, seawater eutrophication emerged as the dominant factor, highlighting the high sensitivity of marine ecosystems to water-quality degradation. By 2020, DEM once again became the leading driver. On one hand, topography determines the spatial distribution of marine development activities, with nearshore areas experiencing denser development and stronger human disturbance, while offshore areas witness a gradual decrease in development activities, resulting in weaker disturbances. On the other hand, topography influences benthic biodiversity and ecosystem stability by modulating processes such as flow velocity and sediment transport (Jones and Frid, 2009). From 2010 to 2015, the seawater eutrophication index became a major driver of marine habitat quality. Between 2010 and 2015, increasing coastal development activities in Jiangsu were accompanied by insufficient awareness and management measures, leading to the accumulation of nutrients, which significantly impacted the marine habitat quality in the study area. Although short-term eutrophication can increase marine primary productivity, the decomposition of algae leads to a decline in dissolved oxygen levels, eventually forming hypoxic zones that cause fish and other organisms to die, disrupting the balance of the habitat (Diaz and Rosenberg, 2008). This result highlights the non-linear response of marine habitat quality to human activities (Hunsicker et al., 2016). While marine habitat quality is predominantly governed by natural factors, when the intensity of human activities exceeds a certain threshold, the impact of human activities may surpass that of natural factors, becoming the primary driver of changes in marine habitat quality. In terms of interactive effects, current velocity consistently played a critical regulatory role. Flow velocity not only directly impacts sediment transport and benthic habitats but also significantly regulates habitat quality through complex interactions with other factors, such as sediment type and eutrophication indices (Luo et al., 2022). Marine habitat quality is primarily driven by natural factors, which can be attributed to the three-dimensional dynamics of marine ecosystems and the relatively low intensity of human activity interference. Compared to terrestrial ecosystems, marine ecosystems exhibit greater openness, allowing marine organisms to demonstrate strong resilience to localized disturbances (Carr et al., 2003). Additionally, human activities have a relatively short history in marine development, and the pressure exerted by human activities on marine ecosystems is inherently fragmented (Halpern et al., 2008). However, once human activities exceed a certain threshold, their impacts can surpass natural factors and become the dominant driver of marine habitat degradation.

Overall, the response of coastal habitat quality to changes in the natural environment and human activities exhibits land-sea differences. These differences are primarily attributed to variations in the intensity of human activities between land and sea, as well as differences in habitat sensitivity to external disturbances. The land-sea divergence in the mechanisms influencing coastal habitat quality provides a scientific basis for formulating targeted coastal ecological protection and restoration policies. For terrestrial habitats, more stringent environmental protection measures can be implemented in areas with high population density and rapid economic development, such as restricting the expansion of construction and agricultural land, promoting urban greening, and enhancing pollution control. In contrast, for marine habitats, greater attention should be paid to the impact of changes in natural factors, such as topography and hydrodynamic conditions, on habitat quality. Additionally, the non-linear response of marine habitat quality to human activities should also be considered.

Current research has certain limitations. Firstly, while this study has explored the land-sea differences in the characteristics, spatiotemporal changes, and driving mechanisms of coastal habitat quality, the interactions between land and sea were not investigated in depth. Land-sea interactions play a significant role in coastal ecosystems, with impacts extending beyond changes in habitat quality to include material cycling, energy flow, and linked ecological processes. Future research should place greater emphasis on the dynamic processes of land-sea interactions and their combined effects on coastal habitat quality.

Secondly, in terms of driving mechanisms, although this study has quantitatively analyzed the influence of various driving factors, the nonlinear interactions between these factors still require further exploration. Particularly under the joint influence of human activities and natural processes, there may be synergistic or antagonistic relationships between different factors. These complex interactions were not fully reflected in this study, and future research could employ advanced techniques such as deep learning to conduct a more detailed analysis.

In conclusion, future studies need to more comprehensively integrate the land-sea interaction processes and multidimensional driving mechanisms to enhance the understanding of the evolutionary patterns of coastal habitat quality. Additionally, research should consider longer time scales and finer spatial resolutions to better reveal the dynamic characteristics of coastal ecosystems, thereby providing more robust support for scientific management and sustainable development.