Abstract

Labor consumption of phytomass measurements on sample plots hinders the extrapolation of point data to areas comparable to commercial lands or landscape units. Vegetation indices calculated from satellite images are usually considered indicators of green phytomass; they are used for its areal estimates. The task of this study is to establish the informativity of the normalized differential vegetation index (NDVI) depending on the fractional structure of living and dead aboveground phytomass, seasonal dynamics of the biological cycle, hydrothermal conditions, and landscape position. The results of monthly measurements of aboveground phytomass fractions on 13 sites in phytocenoses of forest feather grass and feather grass formations in the Burtinskaya steppe (Orenburg Nature Reserve) from May to September 2015–2020 were used. The NDVI values were calculated for each period from Landsat satellite images at all sites. Hypotheses about geobotanical, hydrothermal, phonological, and landscape factors of NDVI informativity were tested by calculating the Spearman correlation coefficients using the dispersion and multiregression analysis. The discrepancy between the seasonal peaks of the NDVI and green phytomass is not consistent with the common concept of a direct indicator value of the NDVI. The total live biomass correlates more clearly with the index in June and July; the correlation is lower in the end of the season. The NDVI index turned out to be sensitive not so much to the reserves of green phytomass, but more to the mass and proportion of forbs and the ratio of live and dead phytomass. In late spring and early summer, the NDVI is most closely associated with forbs and, in July, with cereals. The hypothesis about the possibility of screening green mass with standing dead biomass was confirmed, which leads to a decrease in the NDVI despite the preservation or growth of green phytomass. The NDVI may underestimate the real green phytomass if there is a sharp increase in the mass of standing dead biomass, usually in the second half of summer and early autumn. The NDVI more adequately reflects the state of the aboveground phytomass of steppe communities that have not been exposed to fires for a long time when compared with burned communities and fallows.

中文翻译：

---

作为草原地上植物群信息源的归一化植被指数评价

摘要

样地植物质量测量的劳动力消耗阻碍了将点数据外推到与商业土地或景观单位相当的区域。根据卫星图像计算出的植被指数通常被认为是绿色植物质量的指标；它们用于面积估计。本研究的任务是根据活体和死体地上植物群的分数结构、生物循环的季节动态、水热条件和景观位置，建立标准化差异植被指数（NDVI）的信息性。使用 2015 年 5 月至 2020 年 9 月对 Burtinskaya 草原（奥伦堡自然保护区）森林羽毛草和羽毛草地层的植物群落 13 个地点的地上植物质量分数进行月度测量的结果。根据所有站点的 Landsat 卫星图像计算每个时期的 NDVI 值。通过使用离散度和多元回归分析计算 Spearman 相关系数，检验了有关 NDVI 信息性的地植物学、热液、语音和景观因素的假设。 NDVI季节峰值与绿色植物群的差异不符合NDVI直接指标值的普遍概念。 6月和7月总活体生物量与该指数的相关性更为明显；季节末相关性较低。结果表明，NDVI 指数对绿色植物群的储量不太敏感，而对非禾本科植物的质量和比例以及活植物群和死植物群的比例更敏感。春末夏初，NDVI 与禾本科植物的相关性最为密切，而在 7 月，NDVI 与谷物的相关性最为密切。关于用静止的死生物量筛选绿色植物体的可能性的假设得到证实，尽管绿色植物体保存或生长，但导致 NDVI 下降。如果死生物量急剧增加（通常是在夏末和初秋），NDVI 可能会低估真实的绿色植物量。与被烧毁的群落和休耕地相比，NDVI更充分地反映了长期未遭受火灾的草原群落的地上植物群状况。