Change Analysis of Spring Vegetation Green-Up Date in Qinba Mountains under the Support of Spatiotemporal Data Cube

Abstract

In recent decades, global and local vegetation phenology has undergone significant changes due to the combination of climate change and human activities. Current researches have revealed the temporal and spatial distribution of vegetation phenology in large scale by using remote sensing data. However, researches on spatiotemporal differentiation of remote sensing phenology and its changes are limited which involves high-dimensional data processing and analysing. A new data model based on data cube technologies was proposed in the paper to efficiently organize remote sensing phenology and related reanalysis data in different scales. The multidimensional aggregation functions in the data cube promote the rapid discovery of the spatiotemporal differentiation of phenology. The exploratory analysis methods were extended to the data cube to mine the change characteristics of the long-term phenology and its influencing factors. Based on this method, the case study explored that the spring phenology of Qinba Mountains has a strong dependence on the topography, and the temperature plays a leading role in the vegetation green-up date distribution of the high-altitude areas while human activities dominate the low-altitude areas. The response of green-up trend slope seems to be the most sensitive at an altitude of about 2000 meters. This research provided a new approach for analysing phenology phenomena and its changes in Qinba Mountains that had the same reference value for other regional phenology studies.