为探究全球气候变暖条件[HJ1.8mm]下黑龙江省覆被变化特征，基于1982—2015年GIMMS（global inventory modeling and mapping studies） NDVI3g（normalized difference vegetation index3rd generation）数据，利用趋势分析、Hurst指数法和相关性分析法，分析黑龙江省近34年的植被指数时空变化规律和未来趋势，[JP2]同时结合省内34个气象站点数据，分析归一化植被指数（normalized difference vegetation index，NDVI）与气温、降水及蒸散量的响应特征。结果表明：黑龙江省3种植被覆盖类型的NDVI均呈缓慢上升趋势，其中，林地和草地NDVI增速为0.004/10 a，农田NDVI增速为0.002/10 a；植被覆盖的整体变化先升高后降低，其中最大值出现在2014年，最小值出现在1984年；全省植被覆盖空间格局呈东高西低的分布特征，以“大兴安岭—伊春—鹤岗—佳木斯—双鸭山”为界限，西南地区植被覆盖较为丰富，而东北部地区植被覆盖较少；全省植被变化的同向特征强于反向特征，持续性区域和反持续性区域分别占总面积的97.65%和2.35%；植被覆盖变化与气温相关性高于降水，蒸散与植被指数的相关性在年均和生长季差异性较显著，气温是影响生长季植被覆盖变化的主要气候因子。研究结果可为气候变化背景下寒区土地利用/覆被变化研究提供一定的理论支撑。

In order to explore the characteristics of cover change in Heilongjiang Province under global warming.The spatiotemporal changes and future trends of vegetation cover in Heilongjiang Province in recent 34 years were analyzed based on GIMMS NDVI3G data from 1982 to 2015.At the same time,the correlation between vegetation cover and temperature,precipitation and evapotranspiration were analyzed based on the data of 34 meteorological stations in the province.Research results can provide theoretical support for the study of land-use/cover change in cold regions under the background of climate change. Research methods were trend analysis,Hurst index method and correlation analysis method.Trend analysis method used linear regression analysis to calculate the time-varying variables to determine the significance of the interannual changes in variables,and then used slope to reflect its changing trend.Hurst index method was based on current trends and the range of H value to analyze whether future trends of NDVI was consistent with the current,so as to predict the future trends of NDVI.Correlation analysis method was based on Pearson correlation coefficient to analyze correlations of NDVI with precipitation,mean temperature and evapotranspiration,to study climatic factors influence on NDVI. The results showed that:During 34 years,the three types of planting coverage in Heilongjiang Province all showed a slow upward trend.The growth rate of NDVI in woodland and grassland was 0.004/10 a,and that in farmland was 0.002/10 a.The overall change of vegetation cover increased first and then decreased,with the maximum value appearing in 2014 and the minimum value appearing in 1984.The vegetation cover in Heilongjiang Province is high in the east and low in the west.Taking "Daxing′an mountains-Yichun-Hegang-Jiamusi-Shuangyashan" as the boundary,the vegetation coverage is relatively rich in the southwest,while less in the northeast.The same feature of vegetation change in Heilongjiang Province is stronger than the opposite feature,and areas showed persistence and anti-persistence accounted for 97.65% and 2.35% of total area,respectively.In Heilongjiang Province,the correlation between vegetation cover change and air temperature is higher than that of precipitation,and the correlation between evapotranspiration and vegetation index is significantly different in annual and growing seasons.The average temperature is the main meteorological factor affecting vegetation cover change in the growing season. The overall vegetation cover change in Heilongjiang Province showed a slow increasing trend from 1982 to 2015.The rising trend of NDVI in growing season was more obvious than that in annual average,and there was a good consistency in fluctuation of NDVI between annual average and growing season.The spatial heterogeneity of vegetation cover change was obvious,and most NDVI changes were within ±0.1.Vegetation coverage in Daxing′an mountains increased to a high degree;Vegetation cover change is not obvious in high altitude areas;Reduction of NDVI in the southeast of Sanjiang Plain and some areas of Xiaoxing′anmountains is obvious.In different land use types,woodland in mountainous areas is higher than that in plain farmland and urban areas.During the growing season,NDVI in cropland and grassland showed a significant increasing trend,while NDVI in woodland showed a slight decreasing trend.According to the prediction of the future trend of NDVI change,there were many areas where the future change trend of vegetation cover index showed a good trend,and there were also many areas where the future change was uncertain,and vegetation cover index showed a degeneration phenomenon in the southern region.Regions of uncertain about NDVI future changes were mainly distributed in Songnen Plain,central Region and southern Sanjiang Plain of Heilongjiang Province.In different periods,NDVI future variation trend was obviously different,and the spatial variability was obvious.NDVI future variation trend showed continuous degradation and continuous improvement mainly due to changes in geographic location and climate factors.The positive correlation between NDVI and temperature was weak,and the negative correlation between NDVI and precipitation was weaker,so NDVI response to air temperature is more pronounced.Compared with forest land,cities had weaker self-regulation ability,so NDVI located in cities was more responsive to climate factors.NDVI changes were significantly consistent with changes in air temperature and precipitation in the southwestern urban areas of Heilongjiang Province.About NDVI response to temperature and precipitation,spatial heterogeneity was obvious.Especially in the vegetation growing season,its growth was mostly influenced by precipitation.However,in the Suihua area of Songnen Plain,there was no obvious difference in vegetation cover change affected by temperature and precipitation.