欢迎访问林业科学,今天是

林业科学 ›› 2019, Vol. 55 ›› Issue (8): 9-18.doi: 10.11707/j.1001-7488.20190802

• 论文与研究报告 • 上一篇    下一篇

基于MODIS-NDVI的中老缅交界区近16年植被覆盖时空变化特征

李杰, 张军, 刘陈立, 杨旭超   

  1. 云南大学资源环境与地球科学学院 昆明 650504
  • 收稿日期:2017-11-25 修回日期:2019-05-20 发布日期:2019-09-05
  • 基金资助:
    云南省应用基础研究计划项目“中高分辨率遥感数据支持下的橡胶林地信息定量识别方法研究”(2013ZD002);高分辨率对地观测系统重大专项省(自治区、直辖市)域产业化应用项目“面向南亚东南亚地区‘一带一路’云南高分综合应用示范”(89-Y40-G11-9001-15/18)。

Spatiotemporal Variation of Vegetation Coverage in Recent 16 Years in the Border Region of China, Laos, and Myanmar Based on MODIS-NDVI

Li Jie, Zhang Jun, Liu Chenli, Yang Xuchao   

  1. College of Resources Environment & Earth Science, Yunnan University Kunming 650504
  • Received:2017-11-25 Revised:2019-05-20 Published:2019-09-05

摘要: [目的]探讨中国、老挝、缅甸三国交界区2000—2015年植被覆盖的时空分异和演化趋势,为区内植被的科学管理和有效保护提供参考。[方法]基于中老缅交界区近16年MODIS-NDVI时序数据,借助3S技术以及均值、趋势、变异系数、Hurst指数等统计学相关方法,从多层次多角度探究区内植被覆盖的时空格局、演化规律、空间变异、可持续性及未来演化趋势等特征。[结果] NDVI年均最大值和最小值分别出现在2013(0.779 4)和2002年(0.725 9),整体呈增加趋势,增速表现为每10年增加1.05%; NDVI月均值以3和9月为折点,呈"S"形变化特征,其值为0.698 6~0.831 6,并呈上升趋势,月均增长率为0.50%;区内植被覆盖率较高,16年NDVI均值大于0.6的高植被覆盖区占比97.45%,多集中于热带雨林连片分布的山区,低于0.6的区域仅占2.55%,以各国境内主要城市中心及其外围、山区大面积裸地和澜沧江-湄公河流域沿线等区域为主;区内NDVI随海拔增加表现出持续降低趋势,高值区(NDVI≥0.6)多集中于1 500 m以下的中低海拔地区; 2000—2015年,NDVI时间序列呈改善、退化和不变等变化趋势的区域分别占54.06%、15.62%和30.32%,其变异系数为0.024 4~0.468 8,空间分布上表现为较低波动变化区域 > 低波动变化区域 > 中波动变化区域 > 较高波动变化区域 > 高波动变化区域,低波动和较低波动变化区域占比合计78.49%; NDVI时间序列呈退化区域与呈高波动变化区域具有较明显的空间一致性,多集中于城镇、交通要道等建设用地以及山区大面积裸地等地区;未来,区内植被覆盖将延续过去16年变化趋势的区域占57.15%,与过去16年变化趋势相反的区域占41.09%;结合空间变化趋势特征,发现研究区未来将有39.63%、29.83%和28.98%的区域植被覆盖分别向良性、不变和恶性方向发展,1.56%的区域发展趋势不确定。[结论]中老缅交界区植被覆盖整体较好,16年来,区内植被覆盖随时间(年际、月际)变化幅度较小,并在此基础上整体呈现增加趋势,植被发展前景良好;然而,以建设用地和裸地为主的部分区域,其植被覆盖未来将出现退化趋势。区内各国应合理规划经济发展,节约、集约利用土地资源,并因地制宜地开展植树造林、退耕还林还草等生态修复工作,以促进区域生态环境的良性发展。

关键词: 植被覆盖, MODIS-NDVI, 动态变化, 中老缅交界区

Abstract: [Objective] The spatiotemporal differentiation and evolutionary trends of vegetation coverage in the border region of China, Laos and Myanmar (BRCLM) from 2000 to 2015 were discussed in order to provide support for scientific management and effective protection of vegetation in the region.[Method] Based on the time series of MODIS-NDVI in the BRCLM in recent 16 years, GIS,GPS,and Remote Sensing technologies in combination with statistics-related research method such as averaging, trend analysis,variation coefficient, Hurst index were used to investigate, the spatiotemporal pattern, evolution, variation and developmental trend in the future of vegetation coverage in the region.[Result] For NDVI annual mean,the maximum and minimum appeared in 2013 (0.779 4) and 2002 (0.725 9), respectively, it showed an increasing trend as a whole, with the growth rate increased by 1.05% every 10 years. Monthly average of NDVI showed a changing trend of "S" type with turning pointsin March and September. In addition, it ranged from 0.698 6 to 0.831 6 indicating a rising trend, with an average monthly growth rate of 0.50%. The vegetation coverage rate in the area was relatively high. Regions with 16-year average of NDVI larger than 0.6 accounted for 97.45% and most of which were concentrated in the mountainous areas where the tropical rain forest was distributed in succession. On the other hand,the areas with NDVI's average smaller than 0.6 accounted for only 2.55%, in which the main urban centers and peripheries in three countries, large areas of bare land in mountainous areas and along the Lancang-Mekong River Basin were the main areas. NDVI in the regions showed a decreasing trend with the increase of altitude. In addition, we found that the areas with high NDVI values (≥ 0.6) were mostly concentrated in the middle and low altitude areas below 1 500 m. From 2000 to 2015, the regions which NDVI time series showed improvement, degradation and stable accounted for 54.06%, 15.62%, and 30.32%, respectively, variation coefficient of NDVI time seriesranged from 0.024 4 to 0.468 8, and the spatial distribution was as follows:lower fluctuation region > low fluctuation region > medium fluctuation region > higher fluctuation region > high fluctuation region, the total proportion of low fluctuation region and lower fluctuation region was 78.49%. It was found thatregions with degradation of NDVI time series have obvious spatial consistency with those with large fluctuation, which mainly concentrated on construction land such as cities and towns, major traffic roads, and extensive bare land in mountainous areas. In terms of developmental trend in the future, the areas where vegetation coverage continue tobe the same trend as in the past 16 years accounted for 57.15%, and the areas with contrary trend accounted for 41.09%. Combined with the characteristics of spatial evolution, it was found that about 39.63%, 29.83% and 28.98% of the regions would develop in the direction of benign, unchanged and malignant in the future, respectively, while 1.56% of the regional development trend would be uncertain.[Conclusion] The vegetation coverage in BRCLM was generally good. In the past 16 years, the changing range of vegetation coverage in the area was relatively small.Andon this basis, it showed an increasing trend as a whole, the developmental prospects of vegetation were good. However, in the future, vegetation coverage in some areas with construction land and bare land would be degraded. Countries in the region should plan for economic development rationally, economical and intensive utilization of land resources, and carry out ecological restoration such as afforestation, converting the land for forestry and pasture according to local conditions to promote the healthy development of regional ecological environment.

Key words: vegetation coverage, MODIS-NDVI, dynamic change, border region of China, Laos, and Myanmar

中图分类号: