Search

• High spatial resolution data has a higher accuracy in acquiring SOS. • The SOS changes are mainly affected by spring temperature and winter precipitation. • At lower elevations, advancing SOS increased spring and summer growth. • At higher elevations, delayed SOS leads to decreased spring growth and increased summer growth. • The effect of SOS on vegetation growth weakens from the spring to summer seasons. Vegetation spring phenology in mountains has undergone profound changes due to climate change. In this paper, which made use of the Harmonic Analysis of Time Series (HANTS) smoothing algorithm and logistic models, the start of the vegetation growing season (SOS) inferred from MODIS datasets with different spatial resolutions from 2001 to 2018 was extracted. The MODIS data closest to the ground-truth observation data were then selected to reflect spring phenology. Combined with temperature and precipitation data, the spatio-temporal patterns and elevation-dependent change mechanisms of the SOS and its legacy effect on vegetation growth in the mountains of northwest Mongolia was examined and the following results were obtained: (1) Compared with the MOD13A1 and MOD13A2 datasets, the high spatial resolution MOD13Q1 dataset provided stronger correlations and the lowest errors with ground-truth observation data. (2) In this study, the SOS usually occurred between 120 and 152 day-of-year (DOY) with the earlier SOS detected in the low-elevation areas. Later SOS occurred mostly in the higher elevation areas. (3) The spatio-temporal variations in the SOS are mainly influenced by both spring temperature and winter precipitation, with increasing precipitation in winter tending to delay the SOS, while increasing spring temperature tended to advance the SOS. (4) At elevation gradient, the spring temperature showed a significant negative effect on SOS at all elevations, except between 1037 m and 1137 m and 3337 m to 3437 m above sea level (ASL), with a significant positive correlation between SOS and winter precipitation between 1337 m and 2937 m ASL. It is expected that spring temperature is closely linked to the heat accumulation of plants, while winter precipitation indirectly affects the temperature in spring. (5) In lower elevation areas (below 1437 m ASL), the SOS was found to be negatively correlated with spring and summer growth, with advancing SOS leading to increased vegetation growth during spring and summer. In higher elevation areas (above 1437 m ASL), the SOS was significantly negatively correlated with spring growth and positively correlated with summer growth. Hence, delayed SOS leads to decreased spring growth and increased summer growth. Also, we found the effect of SOS on vegetation growth gradually weakened from spring to summer. These findings can help to determine the mechanisms and functional consequences of spring phenological changes in mountain regions. [ABSTRACT FROM AUTHOR]