Nearly 70% of the Earth’s surface is dominated by low-stature plants, including grasslands, shrublands, Arctic tundra and savanna ecosystems. These ecosystems are vulnerable to changes in climate but the vegetation biomass is difficult to measure with existing observational techniques.
Photogrammetry is the science of making measurements from photographs. Low-cost drones equipped with cameras make it easier than ever before to survey landscapes. Using computer vision we can reconstruct 3D models (point clouds) from these photographs.
These point clouds can enable detailed measurments of subjects, including small differences in biophysical structures, and enable the estimation of plant biomass across entire landscapes. However, photogrammetric measurements are sensitive to the ways that data are collected, processed and analysed, making it hard to compare results produced by different workflows.
To overcome these differences, I developed a protocol for drone photogrammetry for biomass measurement in low-stature ecosystems and led a global collaboration to establish how plant volume relates to aboveground biomass. We conducted photogrammetric surveys over 880 harvest plots, sampling 51 species across four continents.
Our coordinated photogrammetric approach revealed consistent, accurate, and surprisingly linear relationships between biomass and mean canopy height across low-stature species. Standardised protocols are essential to this success, ensuring performance as good as in situ approaches whilst enabling efficient sampling over larger extents than can be sampled using conventional approaches.
We found that canopy height measured with drone photogrammetry is a strong predictor of aboveground biomass, but that the relationships differ systematically among plant functional types.
We also found photogrammetrically-determined canopy heights are sensitive to wind speed but not to sun elevation during the surveys.
Biomass is a key biophysical parameter for quantifying ecosystem responses to changes in climate and herbivory, but knowledge of biomass change in low-stature ecosystems is hindered by current observational approaches. We showed how systematic and standardised drone photogrammetry can transform capacities for observing ecosystem change in understudied but vulnerable ecosystems. For more information, see:
Cunliffe et al. Drone-derived canopy height predicts biomass across non-forest ecosystems globally (Preprint on bioRxiv, DOI:10.1101/2020.07.16.206011).
Cunliffe, A. and K. Anderson (2019) Measuring Above-ground Biomass with Drone Photogrammetry: Data Collection Protocol, Protocol Exchange, doi:10.1038/protex.2018.134.
Cunliffe et al. (2016). Ultra-fine spatial resolution landscape-scale monitoring of dryland vegetation structure with drone-acquired structure-from-motion photogrammetry. Remote Sens. Environ., 183:129-143. DOI:10.1016/j.rse.2016.05.019. Download PDF
Andrew Cunliffe 