The interconnected nature of the United Nations' Sustainable Development Goals (SDGs) presents a challenge: actions taken to advance one goal can positively (synergy) or negatively (trade-off) impact others. While previous research has explored SDG interactions, effective strategies for managing these trade-offs and leveraging synergies remain under-researched. This study addresses this gap by quantitatively investigating the dynamic and spatial variations in SDG interactions within China, a nation with readily available data and a history of implementing SDG-related policies. The study aims to understand the offsetting effects of shifting synergies and trade-offs, characterize the spatio-temporal patterns of closed loops (virtuous or vicious cycles) within SDG networks, and analyze the dynamics of active-passive relationships among SDGs. Understanding these dynamics in a rapidly developing nation like China offers valuable insights for policymakers globally.

Existing literature on SDG interactions falls into two main categories: qualitative assessments based on expert knowledge or official documents, and quantitative correlation analyses employing methods like Pearson's correlation, Spearman's correlation, multiple factor analysis, and autoregressive correlation. These studies have highlighted the variability of SDG interactions based on factors such as national income, level of sustainable development, population groups, and geographic regions. Some studies have even attempted to investigate causal relationships between SDGs using expert opinion, Granger causal analysis, system models, statistical structure learning, and physics-inspired approaches. However, a critical gap exists in understanding the dynamic, offsetting effects of changes in SDG interactions over time and the spatial variations of these changes. The existing literature also largely neglects the fine-scale spatio-temporal dynamics of closed loops and active-passive relationships in SDG causal interaction networks. This study aims to fill these gaps by employing a more sophisticated methodology.

To analyze the causal interactions among SDG indicators in China's 31 provinces (excluding Hong Kong, Macau, and Taiwan) from 2000 to 2020, the researchers employed a spatio-temporal geographically weighted regression (GTWR) model. Data on 86 indicators representing 80 targets and 16 SDGs (excluding SDG14 due to data limitations) were collected from the National Bureau of Statistics and Statistical Yearbooks. These indicators were categorized as positive, negative, or moderate, then rescaled to a 0-100 scale for comparability. Multicollinearity was addressed using the variance inflation factor, resulting in a final dataset of 65 indicators. The GTWR model was used to determine the causal direction and intensity between indicator pairs for each province and year, creating directed weighted networks for both synergies and trade-offs. Network analysis techniques, including the calculation of betweenness centrality and the identification of closed loops (cycles) using a simple cycle algorithm, were applied to the resulting networks. Active-passive relationships were identified by comparing out-degrees (influence exerted) and in-degrees (influence received) of SDG indicators, calculated as the arithmetic average across corresponding indicators for each SDG. The overall change in active and passive relations of each SDG was quantified using a Euclidean distance calculation. The adjusted R-squared was used to evaluate the model fit, with values greater than 0.8 considered credible. A coefficient threshold of ±0.1 was used to define significant synergies and trade-offs.

The study revealed a significant offsetting effect in SDG interactions in China over the past two decades. While the overall numbers of synergies and trade-offs remained relatively constant, substantial shifts occurred between these categories. Approximately 27% of trade-off indicator pairs transitioned to synergies, and about 25% of synergistic pairs became trade-offs. This dynamic interplay highlighted the importance of considering the offsetting nature of these changes. Spatial variations in these transitions were also evident, with different provinces experiencing different patterns of shifting synergies and trade-offs. The analysis revealed the existence of closed loops (virtuous cycles) primarily within the synergy networks, but not in trade-off networks. These virtuous cycles involved specific indicator sets; notably, the study pinpointed high-frequency indicators appearing in virtuous cycles, including those related to responsible consumption and production (SDG12), water management (SDG6), quality education (SDG4), and life on land (SDG15). Analysis of active-passive relations revealed substantial changes over time and space. Several SDGs transitioned from passive to active or vice versa in both synergy and trade-off networks. The study specifically identified several key SDGs (e.g., SDG 1, SDG 4, SDG 7, SDG 11, SDG 13, and SDG 16) which showed notable shifts in their active-passive statuses, indicating changes in their influence and susceptibility to influence from other SDGs.

The findings highlight the complexity of SDG interactions and the limitations of simply counting the number of synergies and trade-offs. The offsetting effect demonstrates the need for dynamic policy interventions that consider the evolving relationships among SDGs. The identification of virtuous cycles provides valuable pathways for effective policy design. Focusing on the high-frequency indicators within these cycles can amplify positive systemic effects. Understanding the active-passive dynamics reveals the vulnerabilities and influential roles of specific SDGs, informing targeted interventions to mitigate trade-offs and enhance synergies. The spatial heterogeneity underscores the need for regionally tailored policies. The results suggest a need for dynamic policy adjustments to address the offsetting effects and capitalize on the potential of virtuous cycles. This implies that policies need to be adaptable and responsive to the changing contexts and interactions among SDGs.

This study provides novel insights into the spatio-temporal dynamics of SDG interactions in China, demonstrating the significant offsetting effects of synergy and trade-off shifts. The identification of virtuous cycles and high-frequency indicators offers a framework for effective policy design to promote synergies and mitigate trade-offs. Further research should explore the impact of specific policies on SDG interactions and investigate the impact of unforeseen events, such as pandemics, on SDG progress. The analytical framework presented can be applied to other countries and scales to facilitate the achievement of the 2030 Agenda.

While the study uses a robust dataset and sophisticated methodology, some limitations exist. Data availability, particularly for SDG 14, restricted the analysis. The GTWR model's reliance on statistical associations may not fully capture the complexities of causal relationships. The study's focus on China might limit the generalizability of findings to other contexts. Future research could explore alternative methods for assessing causal relationships and expand the analysis to include a broader range of SDGs and contexts.