Land use optimization in Ningbo City with a coupled GA and PLUS model

Industrialization and urbanization have resulted in a series of problems such as an excessive intensity of land utilization and an unbalanced structure of land use. Contradictions and conflicts regarding different uses of land are becoming increasingly problematic. The question of how to coordinate the competition among the driving factors of land resources and how to balance and optimize land use types with different functions are urgent problems related to achieving sustainable development. In this study, the traditional model of the land use change process and a machine learning algorithm were integrated to establish an optimal land use allocation model considering macro-quantity control and spatial unit allocation. Taking Ningbo as an example, on the basis of considering the function of land resources and the driving mechanism of land use change, genetic algorithm (GA) and patch-generating land use simulation (PLUS) model were used to optimize the quantitative structure and spatial layout of production‑living‑ecological land (PLEL), respectively. The GA-PLUS model demonstrated strong robustness. The results show that Ningbo covers a total land area of 9816 km², and the rates of increase in the area of production land (PL), living land (LL) and ecological land (EL) in Ningbo from 2010 to 2018 were ‑2.04%, 10.07% and ‑0.69%, respectively. The main problems were the unbalanced distribution of PLEL, the disordered expansion of industrial production land (IPL) and LL, occupying 3.07% of agricultural production land and 0.67% of EL. Guided by Sustainable Development Goals, the construction goals related to the Beautiful China policy and the development planning goals of Ningbo city, the collaborative optimization of the quantitative structure and spatial layout of PLEL in Ningbo can be realized through the GA-PLUS model. In terms of quantitative structure, Ningbo needs to control the total amount of construction land in the future, and improve the proportion of LL and EL, about 13.04% and 51.75%. In terms of spatial layout, it is necessary to build industrial clusters for IPL in coastal areas, improve the suitability of EL in the north of Ningbo, and address the fact that it is difficult for LL to expand due to the influence of the complex terrain in Yuyao and Xiangshan. The results of this study can guide the optimization and development path of the territorial space of Ningbo, and provide technical support for decision-makers to formulate targeted land space planning and realize regional sustainable development.