| Bamboo forest,an important forest type in the subtropical region of China,has strong carbon sequestration ability and strong air purification ability.Besides,bamboo forest grows fast,with high economic benefits.There are large areas of bamboo forests in Fujian Province.At present,it is known that Phyllostachys edulis have strong clonal properties and reproduce fast and asexually through underground bamboo whips.They gradually invade other forest stands and change the forest structure.There is still few research on spatial distribution changes and possible responses of bamboo forest distribution to future climate change.In order to solve the problem of the potential distribution of bamboo forests in the current and future suitable areas of Fujian Province,the MaxEnt(maximum entropy model)was used in this study,and the bamboo forest distribution sample points were collected from the plant specimen database,the vegetation map of Fujian Province.By combing the distribution data,climate,soil and topographic data in Fujian Province,the geographical distribution of the current bamboo forest potential suitable areas in Fujian Province was studied.Then,based on the climate data under the future emission scenarios,the distribution of the bamboo forests in Fujian Province under the two emission scenarios of RCP4.5(2050s,2070s)and RCP8.5(2050s,2070s)forecasting research was studied.In order to solve the problem of how the bamboo forests changed in space in the past,this study took the core area of the Wuyishan National Nature Reserve in Fujian,with the help of high-resolution remote sensing images from year of 2008,2013 and 2019.The results of bamboo forest information extraction were studied to explain the spatial variation characteristics of bamboo forest distribution in pure natural environment in the core area of the reserve in the past 10 years.The main results of this study are as follows:(1)The MaxEnt3.4.1 software was used to model the known distribution sample points of the bamboo forest in Fujian Province in the current period,and the ArcGIS 10.2software was used to visualize the distribution map of the fitness degree output by the model.Under the current climate,bamboo forests in Fujian Province are mainly distributed in the relatively high altitudes of the Wuyi Mountains in the west and the Taimu Mountain-Daiyun Mountain-Bopingling line.The areas with low suitability are not only distributed along the coastline,but also inlaid in inland valleys and basins.The area of non-suitable areas is about 44,000 km~2,accounting for 28.39%of the total area of the province;the area of low-suitability areas is about 36,000 km~2,accounting for 23.23%of the total area of the province.The area of moderately suitable areas is about 47,000km~2,accounting for 23.23%of the province’s total area.The area of highly suitable areas is about 28,000 km~2,accounting for 18.06%of the total area of the province.The most critical environmental factor affecting the distribution of potential suitable areas for bamboo forests in Fujian Province is altitude,and its contribution rate in the modeling process reaches 45%.(2)Using the projection prediction module in the MaxEnt software,import the future climate data in the 2050s and 2070s under the RCP4.5 and RCP8.5 emission scenarios for prediction.The results showed that the bamboo forests in Fujian Province are potentially suitable for the future climate scenarios.The area changes drastically over time.The habitats in the southern of the Wuyi Mountains are degraded greatly,and non-suitable areas expanded.The habitat suitability of the hilly areas of the central mountains increased significantly in the early stage,and formed a concentrated and contiguous distribution trend.Carbon emissions intensified,and the habitats retreated significantly from the southeast to the northwest.The distribution of suitable habitats tended to move toward the coast in the early stage,and then moved to northeast direction in the later stage.In general,climate change will make the habitat suitability of bamboo forests in Fujian Province increase significantly locally,and then significantly decrease as a whole.(3)Based on ENVI,ArcGIS and other software,image preprocessing and bamboo forest information extraction were performed for remote sensing images of three years.The ISODATA unsupervised classification method was used to classify the three periods of remote sensing images.The results showed that the bamboo forest patches in the core area of northeastern Wuyi Mountain had a gradual expansion trend during the study period.In terms of bamboo forest area in the study area,it was 0.82km~2,accounting for0.74%of the total area of the study area in 2008.It was 1.16km~2,accounting for 1.05%of the total area of the study area in 2013.It was 1.56km~2,accounting for 1.41%of the total area of the study area in 2019.According to the comparison between the classification results and the original remote sensing images,the expansion of bamboo forests mainly encroached on the surrounding broad-leaved forests,but the broad-leaved forests expanded to higher altitudes and encroached on the coniferous forests.The phenomenon of expansion threatens are both in broadleaf and coniferous forest habitats.The classification accuracy of remote sensing images in this study is above the more accurate level to meet the analysis needs.To sum up,according to the predictions of the existing climate models,climate change may cause the overall bamboo forest suitability in Fujian Province to experience a process of short-term increase and then a sharp decrease.There is a tendency to move from mountainous and hilly areas towards the coastal direction;and its decreasing process shows the characteristics of shrinking from southwest to northeast.Interpretation of remote sensing images in the core area of Wuyi Mountain shows that the bamboo forest in Wuyi Mountain has expanded in the past 10 years under natural conditions,and mainly threatens the surrounding broad-leaved forest.Future research should pay more attention to the differences in the distribution and changes of bamboo forests at different spatial scales,and monitor the dynamic changes of bamboo forest distribution through the combination of more detailed ecological research and long-term macroscopic remote sensing research. |