B002-0010
Drone-derived canopy height predicts aboveground biomass in non-forest ecosystems across the globe
Monday, 7 December 2020
Poster
Andrew Cunliffe1, Karen Anderson2, Fabio Boschetti3, Hugh A Graham3, Richard E Brazier3, Isla H Myers-Smith4, Thomas Astor5, Matthias M Boer6, Leonor G Calvo7, Pat Clark8, Michael D Cramer9, Miguel S Encinas-Lara10, Stephen Michael Escarzaga11, José M Fernández-Guisuraga7, Adrian Fisher12, Kateřina Gdulová13, Breahna M Gillespie14, Anne Griebel15, Niall P Hanan16, Muhammad S Hanggito17, Stefan Haselberger18, Caroline A Havrilla19, Phil Heilman20, Wenjie Ji21, Jason W Karl22, Mario Kirchhoff23, Sabine Kraushaar18, Mitchell B Lyons24, Irene Marzolff25, Marguerite Mauritz11, Cameron D McIntire26, Daniel Metzen6, Luis A Méndez-Barroso10, Simon C Power27, Jiří Prošek13, Enoc Sanz-Ablanedo7, Katherine J Sauer28, Damian Schulze-Brüninghoff29, Petra Šímová13, Stephen Sitch30, Julian L Smit27, Caitriana M Steele31, Susana Suárez-Seoane32, Sergio A Vargas11, Miguel L Villarreal33, Fleur Visser34, Michael Wachendorf5, Hannes Wirnsberger18 and Robert Wojcikiewicz16, (1)University of Exeter, Exeter, EX4, United Kingdom, (2)University of Exeter, Environment and Sustainability Institute, Penryn, United Kingdom, (3)University of Exeter, Exeter, United Kingdom, (4)University of Edinburgh, School of GeoSciences, Edinburgh, United Kingdom, (5)Universität Kassel, Grassland Science and Renewable Plant Resources, Witzenhausen, Germany, (6)Western Sydney University, Penrith, Australia, (7)University of Leon, Leon, Spain, (8)USDA Agriculture Research Service, Boise, ID, United States, (9)University of Cape Town, South Africa, Cape Town, South Africa, (10)Sonora Institute of Technology, Sonora, Mexico, (11)University of Texas at El Paso, El Paso, TX, United States, (12)University of Queensland, Brisbane, Australia, (13)Czech University of Life Sciences, Prague, Czech Republic, (14)San Diego State University, San Diego, United States, (15)Western Sydney University, Hawkesbury Institute for the Environment, Richmond, Australia, (16)New Mexico State University, Las Cruces, United States, (17)University of Texas at El Paso, El Paso, United States, (18)University of Vienna, Vienna, Austria, (19)Northern Arizona University, Arizona, United States, (20)Agricultural Research Service Tucson, Tucson, AZ, United States, (21)New Mexico State University, Plant and Environment Sciences, Las Cruces, NM, United States, (22)University of Idaho, USA, Moscow, United States, (23)Trier University, Trier, Germany, (24)University of New South Wales, Sydney, Australia, (25)Goethe University Frankfurt, Frankfurt, Germany, (26)University of New Mexico, Albuquerque, United States, (27)University of Cape Town, Cape Town, South Africa, (28)Sul Ross State University, Alpine, United States, (29)University of Kassel, Kassel, Germany, (30)University of Exeter, College of Life and Environmental Sciences, Exeter, United Kingdom, (31)New Mexico State Univ, Las Cruces, NM, United States, (32)University of Oviedo, Oviedo, Spain, (33)USGS, Baltimore, MD, United States, (34)University of Worcester, Worcester, United Kingdom
Abstract:
Non-forest ecosystems, including Arctic tundra, woody savanna, proglacial montane and semi-arid and temperate grassland and shrubland, provide vital ecosystem services including carbon sequestration and forage for grazing, and are highly sensitive to climatic changes. Yet these ecosystems are poorly represented in globally available remotely-sensed biomass products and are undersampled by in-situ monitoring due to spatial and temporal variation. Global change threats emphasise the need for new tools to capture biomass change in non-forest ecosystems at appropriate spatio-temporal scales.
In order to robustly assess the relationships between canopy height derived from drone photogrammetry and aboveground biomass (AGB) across low-stature plant species, we conducted 38 photogrammetric surveys over 741 harvest plots (1500 m2 in total) to sample 50 species through a global site network. Critically, using standardised protocols for sampling, processing and analysing the data enabled direct comparison and data synthesis between different sampling teams.
We found mean canopy height is strongly predictive of AGB across species, with median adjusted R2 of 0.87 (ranging from 0.46 to 0.99) and median prediction error from leave-one-out cross-validation of 3.9%. We found that photogrammetric reconstructions of canopy height were sensitive to wind speed but not sun elevation during surveys. Biomass per-unit-of-height was similar within, but different among, plant functional types. Our standardised photogrammetric approach was generalisable across growth forms and environmental settings, and usually performed as accurately as in situ approaches to biomass estimation.
Drone-photogrammetry enables monitoring of AGB across large spatial extents and has the potential to transform our capacity to observe dynamic and heterogeneous ecosystems. Photogrammetric approaches could provide much needed information to calibrate and validate vegetation models and satellite-derived biomass products, helping us to understand understudied and vulnerable non-forested ecosystems around the globe.
