植物生态学报 ›› 2017, Vol. 41 ›› Issue (9): 972-984.DOI: 10.17521/cjpe.2017.0105

所属专题: 遥感生态学

• 研究论文 • 上一篇    下一篇

腾格里沙漠人工固沙植被演替过程中生物土壤结皮归一化植被指数的变化特征

赵芸1,2, 贾荣亮1,*(), 高艳红1, 周媛媛1,2, 滕嘉玲1,2   

  1. 1中国科学院西北生态环境资源研究院沙坡头沙漠研究试验站, 兰州 730000
    2中国科学院大学, 北京 100049
  • 收稿日期:2016-05-24 修回日期:2017-08-26 出版日期:2017-09-10 发布日期:2017-10-23
  • 通讯作者: 贾荣亮
  • 基金资助:
    国家自然科学基金(41371099、41671210和41530746)和中国科学院“西部之光”人才培养引进计划

Characteristics of normalized difference vegetation index of biological soil crust during the succession process of artificial sand-fixing vegetation in the Tengger Desert, Northern China

Yun ZHAO1,2, Rong-Liang JIA1,*(), Yan-Hong GAO1, Yuan-Yuan ZHOU1,2, Jia-Ling TENG1,2   

  1. 1Shapotou Desert Research and Experiment Station, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou 730000, China

    2University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2016-05-24 Revised:2017-08-26 Online:2017-09-10 Published:2017-10-23
  • Contact: Rong-Liang JIA

摘要:

生物土壤结皮(简称结皮)影响干旱沙区生态系统归一化植被指数(NDVI)。该文基于空间代时间的方法, 利用地物光谱仪采集腾格里沙漠不同始植年代(1956、1964和1973年)固沙植被区发育的5种优势结皮的NDVI数据, 分析了固沙植被演替过程中结皮NDVI的变化特征及其对降水和温度变化的响应规律, 并通过与固沙区主要地物进行比较, 评估了结皮NDVI在该沙区生态系统NDVI中的作用。结果表明: 1)随着固沙植被演替, 结皮NDVI逐渐增加。不同种类结皮NDVI相比较, 土生对齿藓(Didymodon vinealis)结皮>真藓(Bryum argenteum)结皮>混生结皮>地衣结皮>藻结皮。2)结皮NDVI受降水量、气温及二者间的交互作用影响显著, 且具有明显的季节差异。结皮NDVI与降水量及其覆盖土壤浅层含水量均呈显著线性正相关关系, 并且结皮NDVI对水分的敏感性随固沙植被演替而逐渐增加。结皮NDVI总体与日平均气温呈显著线性负相关关系, 而与结皮表面温度呈显著指数负相关关系, 并且结皮NDVI对温度的敏感性随固沙植被演替逐渐增加。结皮NDVI对温度变化的敏感性春季高于夏季, 对水分变化的敏感性夏季高于春季。3)春季降水后, 藓类结皮NDVI显著高于油蒿(Artemisia ordosica)、花棒(Hedysarum scoparium)、柠条(Caragana korshinskii)等灌木及裸沙; 夏季降水后, 结皮NDVI显著低于灌木。若考虑结皮较高的盖度, 春、夏季结皮NDVI对固沙区系统NDVI的贡献率分别为90.01%和82.53%, 均超过灌木(春季9.99%和夏季17.47%), 并且结皮对固沙区系统NDVI的贡献率随着固沙植被演替而逐渐增加, 而灌木的贡献率逐渐降低。该研究证明了在区域尺度上利用结皮NDVI并结合气象资料区分结皮演替阶段的可行性, 并为干旱沙区在区域尺度上地表生态参量遥感估算的误差分析及结皮遥感监测的时相选择提供了数据基础。

关键词: 归一化植被指数, 生物土壤结皮, 植被演替, 降水, 温度

Abstract:

Aims Biological soil crust (hereafter crust) affects normalized difference vegetation index (NDVI) values in arid desert ecosystems. This study aimed to demonstrate the feasibility of combining crust NDVI values with meteorological data to distinguish the crust successional stage at the regional scale. Meanwhile, the characteristics of crust NDVI could provide the basis for the error analysis of NDVI-based surface ecological parameters estimation in desert ecosystems. We also suggested the optimum periods for crust observation based on the multi-temporal remote sensing images.Methods NDVI values of five types of dominant crusts, three typical sand-fixing shrubs and bare sand were collected by spectrometer in the field. Crusts and shrubs were randomly selected in revegetated areas established in 1956, 1964, and 1973 at Shapotou, which is on the southeastern edge of the Tengger Desert. We used the space-for-time method to study the characteristics of crust NDVI values and their responses to precipitation and temperature during the succession process of artificial sand-fixing vegetation. Additionally, we evaluated the contribution of crust NDVI values to the whole ecosystem NDVI values by comparing the NDVI values of crusts, shrubs and bare sand.Important findings 1) With succession process of the artificial sand-fixing vegetation, the crust NDVI values significantly increased. Among different crust types, we found the following order of NDVI values: Didymodon vinealis crust > Bryum argenteum crust > mixed crust > lichen crust > algae crust. 2) Crust NDVI values were significantly affected by precipitation, temperature and their interaction, and the influences showed significant seasonal differences. Furthermore, we found significantly linear correlations between crust NDVI value and precipitation, and between crust NDVI value and the shallow soil moisture content covered by crust. A significantly negative linear correlation between daily mean temperature and crust NDVI value, and a significantly exponential correlation between the surface temperature of crust and its NDVI value. With the succession process of artificial sand-fixing vegetation, the response of crust NDVI value to precipitation and temperature became more sensitive. In addition, the response of crust NDVI value to temperature was more sensitive in spring than in summer, while that to precipitation was less sensitive in spring than in summer. 3) Moss crust NDVI value was significantly higher than that of shrubs and bare sand after the rainfall event in spring, while shrubs NDVI value was significantly higher than that of crust after the rainfall event in summer. Considering the coverage weights of different ground features in sand-fixing areas, crust NDVI values contributed 90.01% and 82.53% in spring and summer, respectively, to the regional NDVI values, which were higher than those of shrubs (9.99% and 17.47% in spring and in summer, respectively). Additionally, with the succession process of artificial sand-fixing vegetation, crust NDVI values contributed more, while shrubs contributed less to regional NDVI values.

Key words: normalized difference vegetation index, biological soil crust, vegetation succession, precipitation, temperature