Habitat destruction and declining ecosystem service levels caused by urban expansion have led to increased ecological risks in cities, and ecological network optimization has become the main way to resolve this contradiction. Here, we used landscape patterns, meteorological and hydrological data as data sources, applied the complex network theory, landscape ecology, and spatial analysis technology, a quantitative analysis of the current state of landscape pattern characteristics in the central district of Harbin was conducted. The minimum cumulative resistance was used to extract the ecological network of the study area. Optimized the ecological network by edge-adding of the complex network theory, compared the optimizing effects of different edge-adding strategies by using robustness analysis, and put forward an effective way to optimize the ecological network of the study area. The results demonstrate that: The ecological patches of Daowai, Xiangfang, Nangang, and other old districts in the study area are small in size, fewer in number, strongly fragmented, with a single external morphology, and high internal porosity. While the ecological patches in the new districts of Songbei, Hulan, and Acheng have a relatively good foundation. And ecological network connectivity in the study area is generally poor, the ecological corridors are relatively sparse and scattered, the connections between various ecological sources of the corridors are not close. Comparing different edge-adding strategies of complex network theory, the low-degree-first strategy has the most outstanding performance in the robustness test. The low-degree-first strategy was used to optimize the ecological network of the study area, 43 ecological corridors are added. After the optimization, the large and the small ecological corridors are evenly distributed to form a complete network, the optimized ecological network will be significantly more connected, resilient, and resistant to interference, the ecological flow transmission will be more efficient.

Landscape ecological health (LEH) assessment of blue–green space is vital for the management and restoration of the urban environment. At present, existing LEH assessment research has mainly focused on the single measurement of landscape pattern or external ecological service function, ignoring the effect mechanism. Moreover, there is a lack of targeted assessment of urban blue–green space LEH. In this study, we constructed an urban blue–green space LEH assessment framework based on the integration of pattern, process, function and sustainability, and conducted an empirical analysis in Harbin, a megacity in Northeastern China. The results showed that the spatial changes in the four assessment units of landscape ecological pattern, process, function and sustainability were not coordinated in the study area. From 2011 to 2020, the overall condition of blue–green space LEH in the study area improved but still at an unhealthy level, and the spatial difference increased. Grassland, water and wetland suffered from the widespread degradation of LEH in the study area, and the LEH level improvement type had the largest area proportion, and the stabilization type had the smallest. Moreover, based on the spatial autocorrelation analysis, we clarified the LEH spatial correlation characteristics of the study area and proposed targeted optimization suggestions. Our assessment framework will extend the LEH assessment scope and methodology, and the research results can provide significant references for urban blue–green space protection and management.

Background Urban green infrastructure (GI) networks play a significant role in ensuring regional ecological security; however, they are highly vulnerable to the influence of urban development, and the optimization of GI networks with better connectivity and resilience under different development scenarios has become a practical problem that urgently needs to be solved. Taking Harbin, a megacity in Northeast China, as the case study, we set five simulation scenarios by adjusting the economic growth rate and extracted the GI network in multiple scenarios by integrating the minimal cumulative resistance model and the gravity model. The low-degree-first (LDF) strategy of complex network theory was introduced to optimize the GI network, and the optimization effect was verified by robustness analysis. Results The results showed that in the 5% economic growth scenario, the GI network structure was more complex, and the connectivity of the network was better, while in the other scenarios, the network structure gradually degraded with economic growth. After optimization by the LDF strategy, the average degree of the GI network in multiple scenarios increased from 2.368, 2.651, 2.189, 1.972, and 1.847 to 2.783, 3.125, 2.643, 2.414, and 2.322, respectively, and the GI network structure connectivity and resilience were significantly enhanced in all scenarios. Conclusions Economic growth did not necessarily lead to degradation of the GI network; there was still room for economic development in the study area, but it was limited under existing GI conditions, and the LDF strategy was an effective method to optimize the GI network. The research results provide a new perspective for the study of GI network protection with urban economic growth and serve as a methodological reference for urban GI network optimization.

The official establishment of China’s national parks marks a new stage in the construction of China’s ecological civilization system. National parks systematically protect the areas with the richest biodiversity and the most complete ecosystem processes in China. This is beneficial not only for China’s natural conservation work, but also for the world’s response to environmental issues, such as climate change. Based on remote sensing images of land use in the four periods 1990, 2000, 2010, and 2020, this study calculated the land use changes in each national park during the corresponding period. Using the Plus model LEAS module, the driving factors of land use change in the national parks were studied and explored. In addition, the study used the InVEST model carbon storage module, using remote sensing images from different periods and the corresponding carbon pools of each national park as the basic data for model operation, to obtain the carbon storage changes in each national park over the past 30 years. Based on the hotspot analysis function, the hotspot areas of carbon storage changes in the national parks in the past 30 years were determined. Consequently, based on the CARS module of the PLUS model, the carbon storage in Northeast Tiger and Leopard National Park in 2030 was estimated under different scenarios. Research suggested that, except for Sanjiangyuan National Park where grassland is the main land use type, the other four national parks are all dominated by forests, and the expansion and changes in the main land use types were due to human activities. In the past 30 years, the carbon storage in China’s national park ecosystem has mainly shown a trend of first increasing and then gradually decreasing. Based on the changes in carbon storage in the national park, restoration scenarios were simulated for the core protected and generally controlled areas of Northeast Tiger and Leopard National Park. Under the ideal scenario, the highest value of carbon storage would be achieved by 2030, which would be 7,468,250 t higher than that in 2020. The present study provides a reference for the regional management of China’s national parks and further confirms that the implementation of the national park system can enhance China’s ability to achieve carbon peaking and neutrality goals.