New Techniques for Measuring Biodiversity With Remote Sensing

| |

It is becoming increasingly clear that our planet is facing not only rapid climate change, but also habitat and biodiversity loss, which is frequently linked to this change.

The limitation for scientists is monitoring biodiversity is not always straightforward, particularly as many regions are remote or impacts and changes to biodiversity can be more subtle and harder to distinguish without visiting specific places.

Now, new techniques using satellite imagery help to make the observation of plant biodiversity not only easier but also could be used to plan future conservation.

Using imaging spectroscopy to map changes in biodiversity

In a recent Nature Communications article, scientists demonstrated that imaging spectroscopy along with vegetation data collected by the National Ecological Observatory Network (NEON) show that landscape level plant biodiversity, called spectral beta-diversity that is calculated directly from spectral images, is able to capture changes and diversity in plant species in all major biomes found in the United States.[1] 

Figure from "Plant beta-diversity across biomes captured by imaging spectroscopy" showing a map of the United States and the location of NEON sites and a graph showing the breakdown of plant types.
Map showing the NEON sites in the United States used for this study and the major biomes covered. Figure: Schweiger & Laliberté, 2022, CC BY 4.0

Results demonstrate that high canopy density are correlated with large plant-to-pixel size on imagery. New sensors, including from NASA’s Surface Biology and Geology mission and the European Space Agency (ESA) Copernicus Hyperspectral for the Environment (CHIME) program, will shortly be offering new hyperspectral data with hundreds of spectral bands that are available at 30 x 30 m pixel resolution and sub-monthly timescales.

The plan is to prioritize the quantification of vegetation distribution using the new hyperspectral data that should be able to capture a wide variety of vegetation types as well as their health and growth stage. In fact, data on the imagery specifications has been released.[2] 

Recently, the Italian Space Agency’s Precursore Iperspettrale della Missione Applicativa (PRISMA) program also released information and data to assist with biodiversity management. Increasingly, countries are releasing threats faced to vegetation diversity and are seeking new space borne platforms to monitor adverses changes to plant communities.[3]

For these missions and data, imaging spectroscopy will be the most important set of approaches used to monitor biodiversity for vegetation. This will require translating spectral signatures from hyperspectral data into specific species and potentially life stages for plants.

Researchers have been increasingly calibrating and improving techniques such that phylogenetic, functional, and structural facets of canopy heterogeneity are improved and better understood with imagery data.

Mapping spatial variation in biodiversity

One area that will require further research, and potentially the new missions could help address, is how well variation in plant species and diversity could be captured across different biomes, including understanding not only the types of plants present but their numbers and density.

Schweiger and Laliberté were able to show that using vegetation inventories from the National Ecological Observatory Network (NEON), progress now seems to be made in better capturing some of this diversity.

A view of an aspen forest showing yellow and green leaved trees with a hill and a ridge in the background.
Spectra from imagery can be used to track changes in beta-diversity using remote sensing. A view of an aspen forest. Photo: Doug Shinneman, USGS, public domain.

The spatial variation for plant species composition in communities is referred to as beta-diversity; researchers used calibrated data from NEON to better match with spectral signatures from hyperspectral data to show that it is possible to get a range of plant diversity information. The researchers were able to show that in 23 out of the 30 NEON sites an average 47% of total variation in plant inventories could be explained by spectra from imagery.

For alpha diversity, or count of species identified for vegetation plot of defined size, results demonstrate sites with closed canopy, or a leaf area index (LAI) greater than 1, had a strong positive relationships in spectral and plant alpha diversity relative to areas with open vegetation.

In high LAI environments, spectral data can demonstrate the ability to capture numeric plant diversity. Smaller plants in more open areas, however, are more difficult to capture. Nevertheless, beta diversity, or plant species composition, appeared to be better captured in different biomes, including forests, shrub, and grassland environments.

Tracking biodiversity over time

One potential application for conservation is by tracking beta-diversity over time one could use hyperspectral imagery to provide an early warning system of ecosystem change. In other words, the research demonstrates that the new imagery coming on line will be useful to potentially also capture threats made to plant diversity.

While this work was largely done in the United States, most biomes could be utilized in this analysis, suggesting global application. Nevertheless, given the limitations in alpha diversity captured by imagery, the researchers suggest using a variety of different monitoring platforms, including aerial imagery, to capture biodiversity change and demonstrate more detail in specific places. 

Increasingly, plant diversity will come under threat due to a variety of reasons, including climate change affecting the composition of various biomes. Our ability to respond and mitigate change will require timely satellite data as well as methods that best capture plant diversity in different ecological settings.

The new satellites coming on line and techniques developed that capture spectral signals for different plants demonstrate we should be better able to assess the effects of plant diversity change in the near future.


[1]    For more on how spectroscopy using hyperspectral imagery can be used to demonstrate plant diversity and change, see:  Schweiger AK and Laliberté E (2022) Plant beta-diversity across biomes captured by imaging spectroscopy. Nature Communications 13(1): 2767. DOI: 10.1038/s41467-022-30369-6.

[2]    For more on NASA’s new imagery specifications to be used, see here:

[3]    The Italian biodiversity program and satellite information can be found here:


Share this article

Enter your email to receive the weekly GIS Lounge newsletter: