Response And Simulation Of Woody Ornamental Plant Phenology To Climate Change In Changsha

To study the phenological changes of woody ornamental plants and their responses to climate change in the Changsha region under global warming,this study selected phenological data and temperature data of 63 woody plants from 1963 to 2014 for classification,regression analysis,correlation analysis,and optimal time period analysis of plant ornamental periods.The phenological and ornamental period changes of woody ornamental plants from 1963 to 2014 were analyzed,Explored the correlation between phenology and temperature,as well as the temperature sensitivity of phenology.Actual observations of plant phenology in Changsha in 2022 were conducted using statistical models and machine learning for simulation analysis.The main conclusions are as follows:(1)Phenological characteristics of woody ornamental plants in Changsha from 1963 to 2014:The green leaf ornamental period of deciduous woody plants is distributed in spring and summer,with a length of 169 to 275 days.The green leaf ornamental period of evergreen woody plants is throughout the year;The flowering period is distributed throughout the year,with a length ranging from 7 to 77 days;The autumn leaf viewing period is distributed in autumn,with a length of 22-69 days;The withered branches are observed during the winter and early spring seasons,with a length of between 64 and 151 days.The overall leaf expansion,flowering,and late flowering stages of 63 plants were advanced,while the leaf discoloration and late defoliation stages were significantly delayed;The length of the green leaf viewing period is significantly prolonged,while the length of the flowering period,autumn leaf viewing period,and withered branch viewing period is shortened.(2)The response of woody ornamental plant phenology to temperature in Changsha City from 1963 to 2014:The phenological period is significantly correlated with the average temperature during the period before phenological occurrence.The average values of the optimal time periods for 63 plant species to be affected by temperature are 176 days,126 days,113 days,76 days,and 106 days before phenological period,including leaf expansion,flowering,late flowering,leaf discoloration,and late defoliation.The average temperature during the optimal period increases by 1℃,leading to a 4.34 day advance in leaf expansion,0.88 days advance in flowering,1.78 days delay in late flowering,4.63 days advance in leaf discoloration,and 6.12 days advance in late defoliation.(3)Phenological differences in different types of woody ornamental plants in Changsha from 1963 to 2014:Evergreen woody plants have a later flowering start and end stage compared to deciduous woody plants,and a longer flowering period.The leaf spreading period of trees is later than that of shrubs,with early and late flowering periods,while shrubs show a delayed trend.The green leaf viewing period,flowering period,and withered branch viewing period are shorter than shrubs,while the autumn leaf viewing period is longer than shrubs.The optimal time period for evergreen plants to be affected by temperature at the beginning of flowering is longer than that of deciduous plant,the optimal time period for evergreen plants to be affected by temperature at the end of flowering is shorter than that of deciduous plant,and the temperature sensitivity at the end of flowering is smaller than that of deciduous plant;The optimal period of flowering for shrubs affected by temperature is shorter than that of trees,the optimal period of leaf discoloration is longer in negative correlation than trees,and the optimal period of late defoliation is longer than trees.(4)Construction of Plant Phenology Model in Changsha City from 1963 to 2014:A statistical model and seven machine learning algorithm models were constructed based on the observation data of woody plant phenology in Changsha from 1963 to 2014,combined with meteorological data.Test indicators were used to evaluate the model The results showed that the root mean square error values of the statistical model and the Decision Tree Classifier,K Neighbors Classifier,and Random Forest Classifier models in machine learning were small,and the model performed well.(5)Observation and Model Testing of Woody Ornamental Plant Phenology in Changsha City in 2022:The actual observation of 15 plant phenology in 2022 continues the overall trend of phenological changes from 1963 to 2014,with early spring phenology,delayed autumn phenology,and extended green leaf ornamental period.However,there is no significant change in the length of flowering and autumn leaf ornamental periods.Four well performing models were used to simulate the phenological period in 2022,and the simulation results were basically consistent with the observation results.The average RMSE of each phenological period in spring was between 3.5 and 7 days,while the average RMSE of each phenological period in autumn was between 5 and 8 days.