| China is the birthplace of tea and has a long history of cultivation.With the implementation of the "the Belt and Road" strategy,tea,as an important resource for economic and trade exchanges,has an important strategic significance in the foreign economic,trade and cultural exchanges between China and the West.At the macro level,promoting the research progress of tea can help build the national "the Belt and Road" strategy,thus making the research highly strategic.At the micro level,tea,as an important cash crop,can bring higher economic benefits to farmers,which has important economic significance.The pattern of tea planting in China has undergone significant changes in the past 30 years,and many problems have arisen during the development process.In order to promote the sustainable development of the tea industry,research will explore the spatial layout and evolution patterns of tea planting in China over the past 30 years,guiding the healthy and orderly development of the tea planting industry.Previous studies on the evolution of crop pattern in academia focused on analyzing the quantitative change and rate characteristics of crop evolution,and lacked exploration on the evolution of crop spatial distribution.Remote sensing monitoring,geographical spatial analysis and other methods that can carry out crop spatial analysis can not realize the simulation and prediction of future crop patterns.Spatial geographical models can simulate and predict crop patterns in the future based on crop pattern analysis,Therefore,this article is based on the method of geospatial modeling and takes China’s tea planting areas as the research object.At the level of land use pattern simulation,the land use status data of 2005 is used to simulate the land use change in 2010.The simulation accuracy is tested by combining the real land use data of 2010.Based on this,the land use change in 2030 is predicted,and the forest land distribution pattern of the simulation results is extracted.At the simulation level of tea planting pattern,based on the internal land type data of the forest land in 2005,the forest land pattern is set as the control to simulate the internal land type changes of the forest land in 2010,and the accuracy of the simulation results is tested.Finally,based on the internal land type data of the forest land in 2005 and the annual forest land control pattern,the evolution of the tea planting pattern from 2005 to 2030 is simulated,and the evolution characteristics of the forest land and tea planting are analyzed.The research results are as follows:(1)The verification results of the coupling analysis of land use types and driving factors show that the area under the ROC curve of cultivated land is 0.865,the area under the ROC curve of forest land is 0.849,the area under the ROC curve of grassland is0.854,the area under the ROC curve of water area is 0.776,the area under the ROC curve of construction land is 0.917,and the area under the ROC curve of unused land is0.912,which proves that the 12 selected driving factors have a strong ability to explain land use types.The accuracy test between the 2010 land use simulation results simulated by the model and the 2010 land use change status data results shows that the Kappa index is 97.47%,with an overall accuracy of 97.66%.The simulation accuracy is relatively high,so based on the 2005 land use status data,the land use change in the study area in2030 can be simulated.(2)The validation results of the coupling analysis of land types within forest land and driving factors show that the area under the ROC curve of tea forest land is 0.705,the area under the ROC curve of forestland is 0.701,the area under the ROC curve of shrub land is 0.711,the area under the ROC curve of open forest land is 0.665,and the area under the ROC curve of other forest land is 0.824.The results of driving factor analysis show that the regression model established is effective,The selected 22 types of driving factors have strong explanatory power for land types.The constructed tea planting pattern simulation model simulated the changes in land types within the forest land in 2010 and the accuracy test of the distribution data of land types within the forest land in 2010.The Kappa index was 89.02%,with an overall accuracy of 92.99%.Based on the internal land type data in 2005 and controlled by the forest land grid,the tea planting pattern in the study area in 2030 was simulated.(3)Through the analysis of the evolution characteristics of land use patterns in the study area from 2005 to 2030,it was found that the areas of cultivated land,water bodies,and unused land showed a fluctuating decrease trend,while the areas of forest land,grassland,and construction land showed a fluctuating increase trend.The characteristics of land use changes in different periods are different,and the land type transfer relationships in different periods are also different.In terms of changes in the number of forest land areas,the overall area of the study area showed a fluctuating increasing trend from 2005 to 2030.In terms of spatial distribution of forest land changes,the spatial change areas of forest land are mainly concentrated in the western provinces,while the forest land pattern in other regions remains stable.The most significant period of spatial pattern change in forest land was from 2010 to 2020,during which other land types shifted significantly to forest land.The spatial pattern change characteristics of forest land from 2005 to 2010 and from 2020 to 2030 were not significant,and the trend of change was relatively stable.The spatial distribution area and planting area of forest land increased throughout the entire period from 2005 to 2030.(4)Through the analysis of the evolution characteristics of tea planting patterns in the study area from 2005 to 2030,it was found that the scale and area of tea planting increased,but there were differences in growth rates between different periods.The growth rate was the fastest from 2005 to 2010,and the growth rate decreased in subsequent years.The dynamic degree of land use was the highest from 2005 to 2010,and the dynamic degree fluctuated from 2010 to 2030,but remained stable overall.The analysis results of land type transfer indicate that there are differences in land type transfer at different time periods.Forest land and sparse forest land are the main land types that transition to tea forest land,while shrub land is the least land type that transitions to tea forest land.The tea planting areas in the research area have significant spatial distribution differences,with obvious density and dispersion characteristics.The tea planting areas are mainly concentrated in the western region.By analyzing the changes in the spatial center of gravity of the tea plantation in the study area,it was found that from 2005 to 2010,the spatial center of gravity of the tea plantation migrated from north to southwest,with relatively long migration distances.From 2010 to 2015,the spatial center of gravity of the tea plantation began to shift and gradually moved westward.From 2015 to 2030,the spatial center of the tea forest shifted eastward,but there were differences in migration levels.The migration levels were relatively weak in2015 to 2020 and 2025 to 2030,while the migration levels were relatively high in 2020 to 2025. |
PDF Full Text Request