Mapping alpha diversity of plant species using scale effects of remote sensing

Many methods have been proposed to estimate biodiversity using remote sensing. However, most approaches are often limited by ecosystem type, sensor type, and study scale. Although the Spectral Variation Hypothesis has been successfully tested in diverse conditions, a controversial spectral diversity–biodiversity relationship was reported in some studies. The rapid loss of biodiversity forces us to develop more new techniques. This work proposed a novel approach to estimating biodiversity based on the scale effects of remote sensing, applicable to a range of ecosystem and sensor types. This study provides a rigorous mathematical proof of the proposed method and validates it in two distinct ecosystems: grassland and shrubland. The correlation between the scale effects of vegetation indices and in situ plant diversity (Shannon’ H and Species Richness) was analyzed across different spatial scales and sensor types. Results reveal that the scale effect of the vegetation index serves as an effective proxy of plant diversity. Vegetation indices that incorporate near-infrared bands demonstrate superior performance, with the optimal index being the Green Ratio Vegetation Index, which utilizes only near-infrared and green bands. In shrubland, r = 0.609, while in grassland, r = 0.489. The study recommends that for low-resolution images, the pixel size should be aligned with the dimensions of field quadrats. In contrast, high-resolution images should approximate the size of individual plants. A comparison with classical methods reports a good performance for the novel method. Moreover, this method is more effective under high vegetation coverage. Specifically, when the vegetation coverage exceeds 60 %, the correlation coefficient rises to 0.718. These findings suggest that the scale effect of the vegetation index is a valuable tool for monitoring plant diversity. Using this method and long time-series remote sensing data to estimate plant diversity in different periods, ecologists can effectively identify degraded areas of plant diversity and formulate conservation measures in time. •Scale effect of vegetation index is a good proxy of plant diversity.•Vegetation indices that include near-infrared bands perform better.•Compared with Shannon's H, the SEM performs better in estimating species richness.•Under the condition of high vegetation coverage, the SEM is more effective.