Trends in NDVI and Tundra Community Composition in the Arctic of NE Alaska Between 1984 and 2009

Trends in NDVI and Tundra Community Composition in the Arctic of NE Alaska Between 1984 and 2009
复制标题

DOI:
10.1007/s10021-015-9858-9
复制
发表时间:
2015-06-01
期刊:
影响因子:
3.7
通讯作者:
Welker, Jeffery M.
Welker, Jeffery M.
中科院分区:
环境科学与生态学2区
文献类型:
--
作者:
Pattison, Robert R.;Jorgenson, Janet C.;Welker, Jeffery M.

文献摘要

被引文献

相似文献

随着北极生态系统经历地表气温升高,对苔原植被组成的地块级分析表明,苔原群落发生了重要变化,其典型特征是灌木的增加和非维管束物种的减少。同时,NDVI 分析表明北极苔原正在绿化。很少有研究将物种组成和覆盖的地块级趋势与遥感测量相结合,以了解苔原植被动态与 NDVI 之间随时间变化的联系。本研究报告了 1984 年至 2009 年阿拉斯加东北部北极国家野生动物保护区野外样地的物种组成趋势,并将这些趋势与精细和粗尺度的 NDVI 趋势联系起来。在此期间,植物群落组成几乎没有变化。测量的五种苔原类型中没有一种总植被覆盖率有所增加,落叶灌木覆盖率也没有显示出其他地方报道的大幅增加。表面(图)测量的 NDVI 与落叶和常绿灌木组成呈正相关,表明这些功能组对 NDVI 值有很大影响。 NDVI 模型值源自对落叶和常绿灌木组成随时间变化的测量,草丛苔原略有下降,但其他苔原类型没有变化。这一结果表明,这些苔原类型的地表 NDVI 并未随时间变化。细尺度(30 米像素)陆地卫星 NDVI 也没有显示位于永久地块(1992-2009 年)的像素的任何变化。然而,整个研究区域的粗尺度(8 公里像素)AVHRR NDVI 确实有所增加(1988-2007 年)。此外,与 AVHRR 像素匹配相同区域的 Landsat 像素的聚合值也没有显示出随时间的显着变化。尽管 Landsat NDVI 与地面测量的 NDVI 一致,但 AVHRR NDVI 却不一致。 AVHRR NDVI 值显示出现场数据和陆地卫星数据均未出现的增加。这一结果表明,AVHRR 可能表现出 NDVI 的增加趋势,而在一些北极苔原生态系统中,这种趋势在地面上并未出现。这些结果突出表明,在分析北极苔原生态系统对气候变化的响应时,需要将遥感与植物群落组成和 NDVI 的实地测量相结合。
As Arctic ecosystems experience increases in surface air temperatures, plot-level analyses of tundra vegetation composition suggest that there are important changes occurring in tundra communities that are typified by increases in shrubs and declines in non-vascular species. At the same time analyses of NDVI indicate that the Arctic tundra is greening. Few studies have combined plot-level trends in species composition and cover with remote sensing measurements to understand the linkages between tundra vegetation dynamics and NDVI over time. This study reports on trends in species composition for field plots in the Arctic National Wildlife Refuge in NE Alaska from 1984 to 2009 and links these trends to the trends in NDVI at fine and coarse scales. Over this time frame there were few changes in plant community composition. None of the five tundra types that were measured had increases in total vegetative cover, and deciduous shrub cover did not show the large increases reported elsewhere. Surface-(plot) measured NDVI was positively correlated to deciduous and evergreen shrub composition suggesting that these functional groups had a strong influence on NDVI values. Modeled values of NDVI, derived from measures of deciduous and evergreen shrub composition over time, decreased slightly for tussock tundra but did not change for other tundra types. This result suggests that surface NDVI did not change over time on these tundra types. Fine-scale (30-m pixels) Landsat NDVI also did not show any changes for the pixels located at the permanent plots (1992-2009). However, coarse-scale (8-km pixels) AVHRR NDVI across the study area did increase (1988-2007). Furthermore, aggregate values of Landsat pixels matching the same area as AVHRR pixels also did not show significant changes over time. Although Landsat NDVI was consistent with surface-measured NDVI, AVHRR NDVI was not. AVHRR NDVI values showed increases that were in neither the field nor Landsat data. This result suggests that AVHRR may be demonstrating increasing trends in NDVI that are not occurring on the ground in some Arctic tundra ecosystems. These results highlight the need to combine remote sensing with on-the-ground measurements of plant community composition and NDVI in the analysis of the responses of Arctic tundra ecosystems to climate change.