The most common method of mapping grassland biodiversity comes from the
spectral variation (SV) hypothesis. This method assumes that the
variability in the spectral signal detected by the remote sensing
instrument is correlated with biodiversity (Rocchini et al., 2004). This
can be performed with both multi and hyperspectral instruments and can
be used to assess biodiversity across a landscape and within plots from
individual sites (Figure 2).
Additional methods for monitoring both biodiversity and functional
traits include the use of different spectral and vegetation indices, 3D
data such as from LiDAR or structure from motion (SfM), data fusion
between high spatial resolution imagery and that with greater spectral
information, or
spectral
data combined with 3D (Aasan et al., 2015; Gašparović et al., 2019;
Laliberte and Rango, 2011).
For this review, a Google Scholar search was conducted for studies
between 2018 and 2024 with the search terms “remote sensing”,
“grasslands” and “biodiversity”, and a separate search replacing
“biodiversity” with “functional traits”. The publications that
included all the search terms and were directly related to the topic
were examined in further detail and their references checked for
additional publications. This resulted in 37 publications, 20 related to
biodiversity, 15 on functional traits, and two studies with a focus on
both biodiversity and functional traits (Table 1).
Table 1: List of the main grassland biodiversity and functional trait
remote sensing studies used for this review.