Vegetation health monitoring

Owned by Maiwenn Herpe
Last updated: May 31, 2021 by Frederika Mourot
3 min read

Satellite(s)

Sentinel-2, Landsat, VHR.

Monitoring element

Land surface reflectance.

Satellite(s)

Sentinel-2, Landsat, VHR.

Monitoring element

Land surface reflectance.

Description technique

Large scale vegetation health monitoring can be achieved using open access multispectral data with Near Infrared (NIR) bands. Many studies have shown the degree of reflectance of light at near infrared wavelengths is suppressed when vegetation is stressed while reflectance at red wavelengths (e.g. Brown et al., 2006, Frampton et al., 2013).
Multiple bands can be combined to give a range of dimensionless, normalized spectral indices, each of which highlights different types of spectral relationships. The Normalized Difference Vegetation Index (NDVI), which combines the red and NIR bands, has become a standard metric (among a range of others) for vegetation health derived from satellite imagery (Frampton et al., 2013).
Normalizing the bands also has the effect of reducing some of the noise resulting from atmospheric interference, sensor calibration etc. Brown et al. (2006) show that NDVI is relatively robust to these effects, with a strong correlation between a range of sensors when compared across the same global locations over a multi-year period.

Health monitoring using vegetation indices can be achieved by relative or absolute means: by comparing units of like species, age etc., then units which show greater stress relative to peers can be highlighted; absolute values can be used to classify units into classes of performance or compare between multiple observations to highlight localized change.

Accuracy / Resolution

Variable spatial and temporal resolution according to sensors.

Case study

Example of the Indufor Plantation Monitoring output, which benchmarks NDVI values across a forest given age and species information to generate a map of relative vegetation health:

https://induforauckland.users.earthengine.app/view/plantation-monitoring

Benefits

  • Gives a consistent, regional scale overview.

  • Can be refined to areas of significance.

  • Allows for targeted application of ground based inspections.

Limitations

  • Requires imagery to be cloud and shadow free to avoid biasing the results, particularly when comparing results between forest stands.

  • When comparing results over time, care must be taken to account for differences in both the illumination conditions of the capture and the dynamic phenology of vegetation itself.

  • A further issue which can confound spectral analysis of vegetation is topographic shadow caused by low sun angle during the winter months.

  • Care must be taken to account for undergrowth. Where the target vegetation does not have a complete canopy, then undergrowth may influence the spectral values of the pixel. This issue may be reduced somewhat by using observations captured during dry periods, where grasses have browned off and may be more easily separated from target vegetation.

Applicability for Northland

  • Yes.

  • Techniques applying optical data will be limited in coverage and temporal granularity by the persistent cloud cover in the region, particularly during the winter months. Mature cloud-masking techniques are directly available for open access multispectral data (e.g. Landsat and Sentinel-2). When using commercial data, care must be taken to ensure that there is sufficiently cloud free imagery available, as cloud masking is not as mature and ordering a large volume of imagery to ensure complete cloud free coverage between multiple observations can become cost prohibitive.

  • While it is possible to predict the location of topographic shadow by using a digital terrain model and optical observation illumination metadata, the complex topography of the Northland region combined with increased winter cloud cover would likely limit vegetation health analysis to summer months.

Publication references

Addabbo, P., Focareta, M., Marcuccio, S., Votto, C. and Ullo, S.L., 2016. Contribution of Sentinel-2 data for applications in vegetation monitoring.
https://core.ac.uk/download/pdf/80268884.pdf

Brown, M.E., Pinzón, J.E., Didan, K., Morisette, J.T. and Tucker, C.J., 2006. Evaluation of the consistency of long-term NDVI time series derived from AVHRR, SPOT-vegetation, SeaWiFS, MODIS, and Landsat ETM+ sensors. IEEE Transactions on geoscience and remote sensing, 44(7), pp.1787-1793.
https://ieeexplore.ieee.org/document/1645279

Frampton, W.J., Dash, J., Watmough, G. and Milton, E.J., 2013. Evaluating the capabilities of Sentinel-2 for quantitative estimation of biophysical variables in vegetation. ISPRS journal of photogrammetry and remote sensing, 82, pp.83-92.

https://www.sciencedirect.com/science/article/pii/S092427161300107X