The Role of Remote Sensing in Vegetation Studies
What is Remote Sensing?
Remote sensing (RS) can be defined as "the acquisition of information about the state and condition of an object through sensors that do not touch it" (Chuvieco, 2020).
This acquisition is carried out through measuring the electromagnetic energy reflected and emitted by the object(s) in question. Sensors are built to pick up this energy in specific portions of the electromagnetic spectrum (EMS) in order to infer characteristics about the objects being observed. These portions of the electromagnetic spectrum are called bands, and can span any portion of the entire EMS.
Sensors may be ground based like probes or monitoring stations, or they may be attached to aircrafts and drones to gather data from above. A large majority of sensors are space-borne, attached to satellite platforms, because of the ability to repeatedly gather data.
Different materials reflect and absorb wavelengths across the EMS in unique patterns. This pattern of reflection is known as the spectral signature. Variation in spectral signatures allow for differentiation between materials across bands within and outside of the visible portion of the EMS.
RS and Photosynthesis
Physical and chemical properties effect spectral signatures. For vegetation, the same biophysical composition that makes photosynthesis possible can be used to leverage information about plants from spectral signatures.
The light energy from the sun is emitted in spectrums all across the EMS, so the spectral signature for vegetation can be measured by sensors that collect the portions of this energy that are reflected back off of the Earth’s surface.
The electromagnetic energy emitted from the Sun peaks in the visible portion of the EMS. Green plants have evolved to absorb energy in this portion of the EMS very well.
When light strikes a green leaf, chlorophyll pigments that reside in the spongy mesophyll of plant cells absorb visible light energy which can be converted into chemical energy through photosynthesis.
The topmost layer of a leaf is called the cuticular surface. It is a waxy, translucent material that allows for both visible and near-infrared light to diffuse through it without reflecting or absorbing any significant amount of energy. Light energy is either absorbed by pigments within cells, reflected back up, or transmitted through the open spaces between cells down toward other leaves or the ground.
There are two different chlorophyll pigments in green leaves, Chlorophyll a and b which both reside in bodies called chloroplasts which are abundant in plant cells. Both pigments absorb red and blue wavelengths while reflecting green wavelengths which cause most leaves to appear green to human eyes (Knipling, 1970).
Healthy vegetation also reflect near-infrared radiation very highly. Because plants' pigments have adapted to absorb energy in the visible portion of the spectrum, absorbing the massive amount of near-infrared wavelengths would cause the leaf to take in too much energy, heat up, and denature cellular structures within the plant. This strong reflectance in the NIR band makes vegetation easily distinguishable from other surfaces in remotely sensed imagery, as you can see below.
Above is a portion of a 1996 Landsat TM scene in Siem Reap, Cambodia. In the upper right portion of the image sits the Phnom Kulen National forest, and at the center the Ankor Wat Temple, the largest religious monument in the world. Left: true color image, Right: RGB composite in which NIR reflectance is highlighted in red, red band reflectance appears as green, and green band reflectance appears as blue. Utilizing these color composites3, called RGBs, helps to exaggerate features on the land which reflect highly in a particular band. In this case, areas of vegetation are much clearer in the false color composite than the true color composite.
Senescence
Although chlorophyll is the dominant pigment in green vegetation, there are other pigments that have different spectral signatures. The additional reflection within the visible portion of the EMS is overpowered by the reflectance of green light by chlorophyll.
These pigments are only seen when the leaves are under stress and in the fall season when they undergo senescence. This is a process in which the chlorophyll content decreases and other pigments in the leaf become the dominant reflective and absorptive structures.
Carotenes are also present in most green vegetation. When this becomes the dominant pigment, the leaf will reflect highly in the yellow portion of the EMS, making the leaves appear yellow. When anthocyanin dominates, leaves will then reflect red wavelengths and absorb green light, producing a bright red color in the leaf. The process produces the array of colors we see in the fall season as deciduous trees lose their leaves for the season. However, pigment changes can also happen throughout the year due to stressors like drought or disease.
Spectral Signature Comparison
By mapping the typical spectral signatures of healthy vegetation a baseline can be made for comparison. Variations in spectral reflectance can be isolated and compared to the biophysical mechanisms in order to infer what type of stress the vegetation may be under.
Healthy photosynthetically active vegetation is know to reflect shortwave infrared portions of the EMS as well. However, this reflection is due to the water content inside leaf cells rather than physiological factors.
When water content is effected, reflectance in green and near infrared wavelengths are also affected due to chlorophyll production. If there is significantly less reflectance in these portions of the EMS, it can be inferred that the vegetation is under stress likely due to drought (Tucker, 1980).
Variation in spectral signatures can also be refined to very small portions of the EMS, like the red-edge band. This band focuses on the sharp increase in reflection from the red to infrared portion of the EMS.
Changes in this jump from absorption to reflection is directly related to chlorophyll content. Variation in this shift can be linked to both vegetation health as well as the stage of growth a plant is in (Horler et al., 1982).
Just as different materials have different spectral signatures, there is also variation in the spectral profiles of specific plants. Due to both structural and biochemical differences, grass, for example, will have a much different spectral signature than an evergreen tree. There is also variation at the species level, which can be utilized for identification and biogeographical mapping (Shouse et al., 2013; Mureriwa et al., 2016).
Vegetation Indicies
Information about vegetation can also be leveraged through mathematical computations of reflectance values between bands which are used to identify the presence and state of photosynthetically active vegetation. These operations are know as indices, and can help to extract valuable information from remotely sensed images.
The images above are taken from a 2009 study of land-cover change in the Ankor Basin from 1989-2005 (Gaughan et al., 2009). An NDVI was preformed on each image to highlight forested vegetation across the landscape in areas of dark green while bare soil, grasslands, water cover, and built areas are light green to white. The image on the left is from 1989 and the image on the right is from 2005. As the swiper is moved back and forth, the loss of forested areas is clear, especially in those areas surrounding the Phnom Kulen National park at the upper right of each image, as well as the areas surrounding the Ankor Wat temple in the center of each image. Gaughan et al. (2009) found that both urban development and increases in tourism cause the use of land and resources to increase, bringing deforestation and a shift in land use toward urban development and agriculture.
There have been many variations and modifications to this transformation in order to incorporate reflectance values indicative of chlorophyl (MCARI) and water (NDWI) content, as well as adjusting for reflectance interference from the atmosphere (ARVI) or soil (SAVI).
There is an abundant amount of vegetation indices (VIs). The normalized difference vegetation index (NDIV) is a basic transformation that uses the inverse relationship between the red and near infrared reflectance. This is the most widely used VI as it typically provides an accurate assessment of the abundance of healthy green vegetation within a remotely sensed image or set of images.
These VIs are often utilized in multi-temporal studies to map and monitor the presence and change of vegetation over time. Depending on the spatial and temporal resolutions, these changes can be depicted at any scale from local to global.
Remote Sensing Applications
One of the largest benefactors of developments in remote sensing technology is the agricultural industry. Using RS, entire fields can be captured and assessed for health and production, making it much more efficient than traditional in-situ measurements (Rocha et al., 2012).
The images above depict the use of NIR to access the health and extent of varying croplands.
RS techniques can be employed to monitor pests and diseases, water and nutrient content, as well as provide estimations of crop yields. These estimations can also help to conserve resources when they are not needed, helping to make the agricultural industry more environmentally conscious and sustainable (Weiss et al., 2020).
As mentioned earlier, RS can be utilized to differentiate between plant species at finer spatial resolutions. Understanding species composition is extremely important for both conservation and natural land management.
Shouse et al. (2013) showed that RS can be utilized for mapping and monitoring invasive species like bush honeysuckle which can overgrown and harm native species. Kolarik et al. (2020) were able to show that RS can be utilized to delineate change in vegetation structure, which can be used for monitoring changes which could lead to land degradation. RS techniques outlined in both applications are able to provide the consistent and time sensitive imagery that is needed for keeping devastating changes at bay.
The image above depicts the use of drone imagery to identify three species, blackberry, guava and cuban cedar, across an agricultural zone in Santa Cruze in the Galapagos to track and map the progression of these invasive species.
Sustainable Development Goals
Evidence of climate change due to anthropogenic activity has become completely undeniable. This fact has forced the United Nations (UN) to take drastic steps for climate action. In 2015, 17 Sustainable Development Goals (SDGs) were outlined and adopted by all countries that are a part of the UN (Morton et al., 2017).
In order to reach these goals by the date of 2030, monitoring of implementation is necessary. RS can provide vital information about the changes to the Earth's surface that can show progress, or lack there of, in these sustainability targets.
Although there are RS applications that can help to monitor the progress of hitting SDG targets, SDGs 13: Climate Action, and 15: Life on Land are specifically linked to vegetation, and can be extensively monitored through the mechanisms outlined above.
Climate has a direct influence on the location, production, and health of vegetation. The biophysical properties discussed above are affected by climactic characteristics like temperature and precipitation. By utilizing the portions of the EMS that indicate change in plant health and productivity, RS can be used to understand what effects a changing climate has had, through imagery collected over time, as well as what steps can be taken to mitigate further anthropogenic influence.
By using VIs like NDVI in RS imagery over global and local scales, vegetation coverage can be mapped and monitored for use in assessing progress of SDG 13. There has been a significant loss of global forested areas over the past century due to anthropogenic effects, which causes progression of climate change.
Climate change has also been directly linked with global carbon dioxide increase which has a strong greenhouse effect when present in the atmosphere. Because vegetation is a significant carbon sink, meaning it takes in and stores carbon dioxide, vegetation monitoring and protection is vital to halting further global warming. RS can play a significant role in monitoring global carbon uptake in vegetation. RS imagery can also play a role in assessing the more dire areas affected by carbon dioxide concentrations and global warming in which environmental policies related to SDG 13 should be implemented to slow and reverse climate change.
The interactive maps below, provided by the NASA earth observatory, show the drastic changes in global vegetation cover over the past two decades. These
RS techniques have already been extensively employed to investigate the ecosystems on land. For this reason RS plays a crucial role in assessing indicators for SDG 15.
The SDG 15.3 target is focused on combating the degradation of land. Multi-temporal RS imagery databases like that of the Landsat missions can be utilized extensively to estimate degradation over time.
Degradation is assessed through the soil carbon stocks, productivity of vegetation, and land cover change (UN, 2020) . The efficiency with which RS can monitor vegetation production is superior to traditional field methods.
By locating those areas hit the hardest, environmental policy and land management can be implemented to direct resources toward restoration of these degraded landscapes and move closer to reaching SDG 15.3.
The SDG 15.5 and 15.8 targets are both concerned with the preservation of biodiversity. The physiological differences that allow for RS to differentiate between plant species can be utilized to monitor positive and negative changes in the biodiversity of vegetation on a landscape.
SDG 15.5 uses a "red list index" to assess the threat of extinction for varying species, those that are close to extinction are flagged as red and monitored more closely (UN, 2020) . The mapping of vegetation species through RS imagery can provide valuable monitoring services for environmental conservationists in preventing the full extinction of threatened plant species.
SDG 15.8 is specifically geared toward the prevention of invasive species. RS techniques for identifying and monitoring invasive vegetation can help to protect biodiversity, as well as stave off land degradation from overgrowth of harmful invasives.
The role of remote sensing in studying vegetation dynamics is vital. With constant advancement in RS technology and imagery, ecological studies will likely become even more dependent and intertwined with the field of RS. The role of remote sensing in assessing and hitting SDGs will also likely follow suit.
Works Cited
•Kolarik, N. E., Gaughan, A. E., Stevens, F. R., Pricope, N. G., Woodward, K., Cassidy, L., Salerno, J., & Hartter, J. (2020). A multi-plot assessment of vegetation structure using a micro-unmanned aerial system (UAS) in a semi-arid savanna environment. ISPRS Journal of Photogrammetry and Remote Sensing, 164, 84–96. https://doi.org/10.1016/j.isprsjprs.2020.04.011
•Knipling, E. B. (1970). Physical and Physiological Basis for the Reflectance of Visible and Near-infared Radiation from Vegetation. Remote Sensing of Environment, 155-159.
•Lechner, Alex & Foody, Giles & Boyd, Doreen. (2020). Applications in Remote Sensing to Forest Ecology and Management. One Earth. 2. 405-412. 10.1016/j.oneear.2020.05.001.
• Moreira, N., Adam, E., Sahu, A., & Tesfamichael, S. (2016). Examining the Spectral Separability of Prosopis glandulosa from Co-Existent Species Using Field Spectral Measurement and Guided Regularized Random Forest. Remote Sensing, 8(2). https://doi.org/10.3390/rs8020144
•Morton, S., Pencheon, D., & Squires, N. (2017). Sustainable Development Goals (SDGs), and their implementation. British Medical Bulletin, 1–10. https://doi.org/10.1093/bmb/ldx031
•Nel, Anton. (2011). The influence of different winemaking techniques on the extraction of grape tannins.
•Rocha, J., Perdigo, A., Melo, R., & Henriques, C. (2012). Remote Sensing Based Crop Coefficients for Water Management in Agriculture. In Sustainable Development - Authoritative and Leading Edge Content for Environmental Management. InTech. https://doi.org/10.5772/48561
•Shouse, M., Liang, L., & Fei, S. (2013). Identification of understory invasive exotic plants with remote sensing in urban forests. International Journal of Applied Earth Observation and Geoinformation, 21, 525–534. https://doi.org/10.1016/j.jag.2012.07.010
•Szantoi, Zoltan. (2013). REVIEW OF THE USE OF REMOTELY-SENSED DATA FOR MONITORING BIODIVERSITY CHANGE AND TRACKING PROGRESS TOWARDS THE AICHI BIODIVERSITY TARGETS.
•Weiss, M., Jacob, F., & Duveiller, G. (2019). Remote Sensing for agricultural applications: A meta-review. Remote Sensing of Environment.
Xue, J., & Su, B. (2017). Significant Remote Sensing Vegetation Indices: A Review of Developments and Applications. Journal of Sensors, 2017, 1–17.
