近20年我国县域耕地物候变化空间格局研究

SPATIAL PATTERNS OF CROPLAND PHENOLOGY AT COUNTY LEVEL OF CHINA SINCE 2000

关键词:  MODISEVI耕地物候县域尺度趋势变化空间格局

Key words:  MODIS  EVI  cropland phenology  county level  change trend  spatial pattern

摘要:

［目的］耕地物候的时空格局研究可以为生态环境建设、农业生产管理等提供依据。［方法］文章基于2001—2017年全国范围内MODIS EVI时间序列数据，使用Savitzky Golay滤波函数对EVI时间序列曲线进行重建，然后利用动态阈值法提取出了耕地的生长季始期、生长季末期和生长季长度，通过Sen氏斜率计算出了3项物候参数在2001—2017年的变化趋势。［结果］（1）从像素单元来看，全国耕地物候生长季始期变化中有5307%的像元表现为提前趋势； 生长季末期物候变化中有7678%的像元表现为推迟趋势； 生长季长度变化中有7003%的像元表现为延长趋势。（2）从县域单元来看，我国耕地物候的变化趋势以生长季长度延长为主，这类县域在全国县域单元中的占比为642%。（3）从空间格局来看，表现为生长季长度延长趋势的县域主要集中在我国东部三大平原、四川盆地以及东部平原向中部地区的过渡带中。［结论］近20年来我国县域耕地物候生长季长度的延长趋势明显且空间差异显著，这是自然地理要素与人类活动等因素综合作用的结果。

Abstract:

The study of spatial and temporal patterns of cropland phenology can provide a basis for ecological environment research and agricultural production management. Based on the Savitzky Golay filter method, the MODIS EVI time series from 2001 to 2017 was reconstructed. And the key phenological parameters of cropland including start of growing season (SOS), end of growing season (EOS) and length of growing season (LOS), were extracted by using dynamic threshold method. Then the change trends of the phenological parameters from 2001 to 2017 was calculated by using the Sen trend method. The results were found as follows. Firstly, from the pixel level, 53.07% of the pixels had advancing trends in SOS, 76.78% of the pixels had delaying trend in EOS and 70.03% of the pixels had extending trend in LOS. Secondly, from county level, the extending LOS dominated the trends of cropland phenology in China and 64.2% counties had this trend. Finally, the counties having extending LOS were mainly concentrated on the three plains in east China and the Sichuan Basin. So, it can conclude that in the past 20 years, the LOS of cropland in China has obvious extending trend and has significant spatial differences, which is the combined effect of natural environment and human activities.