Multiple Time-scale Changes Of NDVI In The Yangtze River Economic Belt And Their Relationship With Climate And Land Use Change

The Yangtze River Economic Belt is not only one of the most densely developed regions,but also an important ecological function area in China.This region has witnessed dramatic changes in the vegetation because of the complex climate change and frequent changes of land use.Therefore,it is in particular necessary to study the trend and driving factors of vegetation change in the Yangtze River Economic Belt.However,existing research mainly focuses on a single time scale without considering multiple time scales.Based on climate data,land use/cover data,and the Normal Difference Vegetation Index(NDVI),this study adopts EEMD,M-K trend test,and correlation analysis to investigate the characteristics of climate and vegetation change and explore the influences of land use changes on vegetation at multiple time scales in the Yangtze River Economic Belt from 1982 to 2015.The conclusions are as follows:(1)The distribution of NDVI in the growing season from 1982 to 2015 has obvious spatial heterogeneity.The distribution displays a decreasing trend from middle to the east and west and from south to the north.The high values are mainly distributed in mountainous areas where human activities are less frequent.Obvious differences are observed in the NDVI distribution in different landform types.The ranking of the average NDVI values in the growing season is: medium undulating mountain> small undulating mountain> hills> large undulating mountain>tableland> plain.(2)The growing season NDVI in the study area mainly exhibits a 3-year time scale and a long-term trend.It is dominated by a 3-year time scale in the large,medium and small undulating mountainous areas of western Sichuan and Yunnan.The NDVI primarily displays a long-term trend in the tableland,hills and small undulating mountains of eastern Sichuan and Chongqing,and in the plains of northern Anhui and Jiangsu.(3)In the EEMD trend test,18.28% of the trend reversals of NDVI is found significantly increased in the M-K trend test in the study area,and 16.37% is found insignificant changed,which indicates that the EEMD trend test can more accurately identify the part where the trend reversals occur and can further clarify the change trend of vegetation NDVI.(4)In the study area,the variation of precipitation is small while the variation of temperature factors is large during the same period.With the increase of time scales,the correlation between NDVI and meteorological factors increases in the growing season.On the 3-year time scale,the correlation between the monthly mean maximum temperature and NDVI is the highest positive,and the negative correlation between precipitation and NDVI is greater than the positive correlation.On the 6-year time scale,the correlation between vegetation NDVI and precipitation,monthly average temperature and monthly mean minimum temperature is higher than that on the 3-year time scale,but the correlation between NDVI and the monthly mean maximum temperature and the monthly average temperature difference is lower than that on the 3-year time scale,and the positive correlation area is greater than the negative correlation area.On the 14-year time scale,the area ratio of the negative correlation between climatic factors and NDVI increases significantly,approaching that of the positive correlation.The area ratio of the negative correlation between NDVI and the monthly average temperature and the monthly mean minimum temperature even exceeds the positive correlation.In the long-term trend,climate factors are significantly correlated with NDVI in most regions during the growing season.Among them,the area ratio of the negative correlation between precipitation and NDVI is higher than the positive correlation,and temperature factors are mainly positively correlated with NDVI.(5)The most obvious changes in land use / cover are the decrease of a large amount of farmland and the increase of urban and rural residential land in the study area.On the 3-year time scale of NDVI,the changes of land use are the conversion of woodland to unused land,grassland to unused land,and unchanged woodland,grassland,and unused land.On the long-term trend,the land use changes are the conversion of farmland to urban and rural residential land,water to woodland,urban and rural residential land to farmland,urban and rural residential land to water,as well as unchanged farmland and urban and rural residential land.Under the condition that land use remains unchanged,cultivated land and urban and rural residents' land are affected by long-term human activities.The long-term growth trend of vegetation NDVI is as obvious as the short-term fluctuations of 3 years;while forest land,grassland,and unused land are mainly short-term fluctuations The long-term trend is weak.(6)The changed farmland and converted grassland promote vegetation greening.The conversion of farmland and forests into urban and rural residential land leads to vegetation degradation.The conversion of water to grassland and unused land causes severe vegetation degradation,and there is a risk of degradation in the future.Through the analysis of multiple time scales,this study not only reveals the changes of vegetation at different time scales and the characteristics of different trends,but also reveals the impact of climate and land use changes on vegetation changes from different time scales,thereby confirming The importance of multi-time scale analysis for in-depth study of vegetation change and its impact mechanism is provided,and it provides theoretical basis and scientific reference for the ecological and economic sustainable development of the Yangtze River Economic Belt.