Landscape patterns have posed a great impact on the natural ecosystem and food production after land use and urbanization. This study aims to explore the spatial and temporal influence mechanisms of watershed landscape patterns on food production services. The food production services were supplied to maintain food security after landscape pattern optimization. Taking Jiangxi Province as the study area, the spatial distribution of grain production (GP) was measured for five periods in 2000, 2005, 2010, 2015, and 2020; The patch density (PD), the mean patch size (MPS), the patch cohesion index (COHESION), and the landscape shape index (LSI) were selected to measure the landscape pattern of the watershed unit, in terms of shape, area and spatial distribution; A systematic investigation was implemented to explore the influence mechanism of landscape pattern on the GP services using spatiotemporal geographically weighted regression model; According to the spatial heterogeneity of influence mechanism, the landscape zones were divided to regulate the landscape pattern. The results show that: 1) The overall supply of food production services showed a fluctuating upward trend from 2000 to 2020, and then fell back in 2015. There were large differences in the capacity of food production services in various basins. The center of gravity was extended from the Poyang Lake Plain to the southwest. The aggregated distribution was observed in the center of the plains and basins. The GP presented the descending pattern of Poyang Lake Plain > Jitai Basin > Gannan Hills > Northwestern Low Hilly Area. 2) There was a more heterogeneous pattern of landscape during the period of 2000-2020. The zoning control of the landscape was then implemented, according to the influence mechanism. There was a great variation in the direction and degree of the landscape pattern index. The patch density and landscape shape index increased by 5.35% and 1.39%, respectively; While the average patch area and landscape connectivity index decreased by 5.14% and 0.68%, respectively. The landscape pattern tended to be complicated, fragmented, and dispersed. The fragmentation of landscape was roughly presented as the distribution pattern of “high in the plains and basins, low in the mountains and hills; high in the north of Gan, low in the south of Gan”. 3) The spatial and temporal evolution of GP was attributed to the joint action of many factors. Each driving factor on the GP landscape shared the piecewise or bandwise pattern of spatial distribution. Among them, cultivated land area, water density, and water shape posed the positive effects on the GP, whereas, the negative effects were observed in the water area, woodland density, and grassland shape. 4) According to the spatial differentiation of landscape factor coefficients, the study area was divided into four clustering areas, namely, northeast Gan monsoon, Poyang Lake plain, central main production, and southern hilly. Furthermore, regulation suggestions were proposed for these regional landscapes. The finding can also provide a strong reference to enhance the food production services and landscape pattern regulation in Jiangxi watersheds.

Landscape fragmentation has threatened the habitat quality in the context of rapid urbanization. A tradeoff between economic development and ecological protection has been achieved in Ganzhou City, Jiangxi Province, China. This study aims to clarify the driving mechanisms of landscape patterns on habitat quality. Land use data was also collected from 2000 to 2020. The optimal landscape patterns were then proposed for ecological protection and decision-making. An InVEST model was applied to measure the spatiotemporal distribution of the quality of the settlement environment using a transition matrix. Furthermore, the global and local spatial autocorrelation of habitat quality was calculated using the GeoDa software. Additionally, the evolution of habitat quality was constructed to balance the LISA time paths and spatiotemporal transition dynamics using scatter plots. The spatiotemporal differentiation of landscape patterns was explored with a 10 km×10 km grid as the measurement unit. A systematic evaluation was implemented to determine the impact of various indices on the habitat quality at the class of land landscape on both time and space using the geographically and temporally weighted regression (GTWR) model. Finally, the optimal threshold intervals were established to regulate regional landscape patterns. Some insights were offered into the interplay between landscape configurations and habitat quality at different spatial scales over time. The results were as follows. The spatial distribution of habitat quality was characterized by 'high in the periphery, low in the center' at an overall high level from 2000 to 2020, indicating positive and synergistic spatial dynamics. There were remarkably stable landscape patterns with the human-driven land expansion to selectively encroach on non-forested and non-aquatic areas. A better agreement was also found on the spatial increment in water bodies, indicating a complex and variable interaction between anthropogenic and natural landscape elements. There was a significantly spatiotemporal non-stationarity in the relationship between habitat quality and landscape patterns. The Index of Interspersion and Juxtaposition (IJI) revealed that the mixed distribution of artificial land, forests, and water bodies was utilized to mitigate the negative impacts of urban expansion on habitat quality. Optimal threshold intervals were identified for the various indices of landscape patterns for better habitat quality in the study area. The landscape configurations had the promising potential to support ecological conservation. The relationship between landscape patterns and habitat quality greatly contributed to sustainable development and integrated management under environmental challenges. The key influencing factors on the habitat quality were determined from the perspective of landscape pattern reconstruction. The findings can provide a strong reference to maintaining a stable ecosystem and sustainable development.