A Comprehensive Analysis Utilizing Long-Term Dataset
Hanchen Duan,Beiying Huang,Shulin Liu,Jianjun Guo,Jinlong Zhang
DOI: https://doi.org/10.3390/ijgi13120457
IF: 3.4
2024-12-18
ISPRS International Journal of Geo-Information
Abstract:The Qinghai–Tibet Plateau (QTP) is crucial for global climate regulation and ecological equilibrium. However, the phenomenon of global climate warming has increased the frequency of extreme weather events on the QTP, exerting substantial effects on both regional and global ecological systems. This study utilized long-term series NDVI and extreme climate indices to comprehensively evaluate the impact of extreme climatic changes on diverse vegetation types within the QTP. A variety of analytical methodologies, including trend analysis, a Mann–Kendall test, correlation analysis, and random forest importance ranking, were employed in this study. These methodologies were applied to investigate the distribution patterns and variation trends of diverse vegetation types and extreme climate indices. This comprehensive approach facilitated a detailed analysis of the responses of different vegetation types to interannual variability under extreme climatic conditions and enabled the assessment of the impact of extreme climate indices on these vegetation types. The findings have the following implications: (1) Except for forests, the annual NDVI for overall vegetation, meadows, steppes, deserts, and alpine vegetation in the QTP exhibits a significant upward trend (p 0.05). Substantial increases in vegetation were predominantly detected in the central and northeastern regions of the QTP, while significant decreases were mostly observed in the southeastern and western regions. The area exhibiting significant vegetation increase (38.71%) considerably surpasses that of the area with a significant decrease (14.24%). (2) There was a statistically significant reduction (p < 0.05) in the number of days associated with extreme cold temperature indices, including CSDI, DTR, FD, ID, TN10p, and TX10p. In contrast, indices related to extremely warm temperatures, such as GSL, WSDI, SU25, TN90p, TNn, TNx, TX90p, and TXx, exhibited a statistically significant increase (p < 0.01). The pronounced rise in minimum temperatures, reflected by fewer cold days, has notably contributed to climate warming. Although extreme precipitation events have become less frequent, their intensity has increased. Notable spatial variations in extreme precipitation were observed, although no consistent changing pattern emerged. (3) The annual NDVI for non-forest vegetation types showed a significant negative correlation with most extreme cold temperature indices and a significant positive correlation with extreme warm temperature indices. A significant positive correlation (p < 0.05) between annual NDVI and extreme precipitation indices is found only in steppe and desert ecosystems, with no such correlation observed in other vegetation types. Both correlation analysis and random forest methodologies underscore the impact of extreme climate indices on vegetation variations, with the random forest model exhibiting superior capability in capturing nonlinear relationships. In conclusion, global climate change is projected to result in a heightened frequency of extreme warm events. Although these conditions might temporarily enhance vegetation growth, they are also associated with numerous detrimental impacts. Therefore, it is imperative to enhance awareness and take proactive measures for early warning and prevention.
geography, physical,remote sensing,computer science, information systems