Optimizing green-adaptive green infrastructure in response to extreme climate changes: insights from a hybrid SEM-ANN...

Despite the potential of Green-Adaptive Green Infrastructure (GAGI) to increase the provision of ecosystem services, and to mitigate urban climate risks while maintaining biodiversity, there is a critical research gap in the empirical identification of the context specific factors in developing nations. Therefore, this study examine the critical barriers, drivers and potential strategies to optimize GAGI by employing satellite imagery and survey data of 1232 respondents from Pakistan. The research rationale stems from the need to align urban expansion with ecosystem resilience. By employing remote sensing techniques on satellite imagery data, land use land cover analysis highlights the significant urban expansion in the study area. For the empirical analysis, a deep learning-based hybrid structural equation modeling-artificial neural network analysis was used to capture linear and complex non-linear relationships. Findings reveal that integrated urban transformation, socioeconomic equity, governance, and infrastructural barriers were critical barriers to implementing GAGI. Among the drivers, economic development and innovation, environmental sustainability, and infrastructure integration and efficiency are the main drivers of GAGI implementation. Moreover, education and R&D support, GAGI incentives, government standards and regulations, publicity programs, and awareness are the potential strategies for implementing GAGI in Pakistan. Specifically, sensitivity analysis of artificial neural network identifies governance and infrastructural barriers, economic development and innovation, and incentives towards GAGI as the most influencing factors. The study highlights the need for GAGI and provides a reference for decision-making on GAGI implementation and landscape transformation in response to extreme climate changes.