欢迎访问《草业学报》官方网站,今天是 分享到:

草业学报 ›› 2023, Vol. 32 ›› Issue (8): 1-13.DOI: 10.11686/cyxb2022370

• 研究论文 •    

2000-2020年内蒙古草地植被覆盖度时空变化及趋势预测

张慧龙(), 杨秀春(), 杨东, 陈昂, 张敏   

  1. 北京林业大学草业与草原学院,北京 100083
  • 收稿日期:2022-09-15 修回日期:2022-10-10 出版日期:2023-08-20 发布日期:2023-06-16
  • 通讯作者: 杨秀春
  • 作者简介:E-mail: Yangxiuchun@bjfu.edu.cn
    张慧龙(1997-),男,河北张家口人,在读硕士。E-mail: huilongzhang@bjfu.edu.cn
  • 基金资助:
    国家自然科学基金(41571105)

Spatio-temporal changes in grassland fractional vegetation cover in Inner Mongolia from 2000 to 2020 and a future forecast

Hui-long ZHANG(), Xiu-chun YANG(), Dong YANG, Ang CHEN, Min ZHANG   

  1. School of Grassland Science,Beijing Forestry University,Beijing 100083,China
  • Received:2022-09-15 Revised:2022-10-10 Online:2023-08-20 Published:2023-06-16
  • Contact: Xiu-chun YANG

摘要:

草地植被覆盖度(FVC)是反映草地生态状况最直接的指标之一。目前在大区域内构建准确的FVC估算模型,进行长时间序列的动态分析,仍是一个挑战。基于大量地面调查样本,使用2000-2020年MODIS遥感数据、气象数据,通过随机森林模型进行FVC分区建模与结果估算。利用Sen+Mann-Kendall趋势分析法、Hurst指数法等,分析FVC时空变化特征和未来变化情况。研究表明:1)内蒙古草地FVC随机森林模型精度表现为分区优于全区,有效地降低了空间异质性的影响。2)内蒙古草地FVC总体上呈东高西低的分布格局,空间差异明显。3)近21年,内蒙古草地FVC总体呈波动上升趋势,年增长率为0.2%·a-1;增长区域面积占比(79.5%)大于减少区域面积占比(20.5%),并且极显著增长和显著增长占比(28.3%)远大于极显著减少和显著减少(1.6%)。4)未来内蒙古草地FVC总体为正持续性发展。预测增长区域(57.6%)多于减少区域(42.4%),其中极显著增长和显著增长占比较高(25.9%),未来草地FVC整体向好发展。

关键词: 内蒙古草地, 植被覆盖度, 随机森林, 时空变化, 趋势预测

Abstract:

Inner Mongolia is an important green ecological barrier in northern China, and grassland degradation in this region is of great concern. Fractional vegetation cover (FVC) is one of the most direct indicators of grassland ecological status. At present, it is still a challenge to build an accurate FVC estimation model for dynamic analysis of a large region over a long period of time. In this study, we used a large ground survey dataset, MODIS remote sensing data, and meteorological data from 2000 to 2020, and applied the random forest model for FVC partition modeling and prediction. The Sen+Mann-Kendall trend analysis method and Hurst index method were used to analyze the spatio-temporal changes in the FVC and its future trends. The main results were as follows: 1) The precision of each partition of the random forest model was better than that of the whole region, which effectively reduced the impact of spatial heterogeneity. 2) In Inner Mongolia, the grassland FVC generally showed a spatial pattern of being high in the east and low in the west, with obvious spatial differences. 3) In the past 21 years, the FVC of grassland in Inner Mongolia showed a fluctuating upward trend overall, the area of increased FVC was larger than the area of decreased FVC, and the magnitude of the extremely significant increase and significant increase was greater than that of the extremely significant decrease and significant decrease. 4) In the future, the grassland FVC in Inner Mongolia will generally improve. The area of FVC growth is larger than the area of FVC decrease, and the area with extremely significant growth and significant growth accounts for a high proportion (25.9%) of the total area. It is predicted that vegetation growth will develop well in the future.

Key words: Inner Mongolia grassland, fractional vegetation coverage, random forest model, spatio-temporal change, trend forecast