Estimating herbaceous aboveground biomass in Sahelian rangelands using Structure from Motion data collected on the ground and by UAV

Taugourdeau, Simon; Diedhiou, Antoine; Fassinou, Cofélas; Bossoukpe, Marina; Diatta, Ousmane; N’Goran, Ange; Audebert, Alain; Ndiaye, Ousmane; Diouf, Alioune; Tagesson, Torbern; Fensholt, Rasmus; Faye, Émile

doi:10.1002/ece3.8867

Cited by 11 publications

(3 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The usefulness of GLI in predicting biomass and green vegetation has been highlighted by previous studies. Taugourdeau et al [ 73 ] found GLI to be among the most important variables when estimating herbaceous above-ground biomass in Sahelian rangelands. In another study, GLI was found to show higher sensitivity in detecting green vegetation [ 74 ].…”

Section: Resultsmentioning

confidence: 99%

Utilizing Spectral, Structural and Textural Features for Estimating Oat Above-Ground Biomass Using UAV-Based Multispectral Data and Machine Learning

Dhakal,

Maimaitijiang,

Chang

et al. 2023

Sensors

View full text Add to dashboard Cite

Accurate and timely monitoring of biomass in breeding nurseries is essential for evaluating plant performance and selecting superior genotypes. Traditional methods for phenotyping above-ground biomass in field conditions requires significant time, cost, and labor. Unmanned Aerial Vehicles (UAVs) offer a rapid and non-destructive approach for phenotyping multiple field plots at a low cost. While Vegetation Indices (VIs) extracted from remote sensing imagery have been widely employed for biomass estimation, they mainly capture spectral information and disregard the 3D canopy structure and spatial pixel relationships. Addressing these limitations, this study, conducted in 2020 and 2021, aimed to explore the potential of integrating UAV multispectral imagery-derived canopy spectral, structural, and textural features with machine learning algorithms for accurate oat biomass estimation. Six oat genotypes planted at two seeding rates were evaluated in two South Dakota locations at multiple growth stages. Plot-level canopy spectral, structural, and textural features were extracted from the multispectral imagery and used as input variables for three machine learning models: Partial Least Squares Regression (PLSR), Support Vector Regression (SVR), and Random Forest Regression (RFR). The results showed that (1) in addition to canopy spectral features, canopy structural and textural features are also important indicators for oat biomass estimation; (2) combining spectral, structural, and textural features significantly improved biomass estimation accuracy over using a single feature type; (3) machine learning algorithms showed good predictive ability with slightly better estimation accuracy shown by RFR (R2 = 0.926 and relative root mean square error (RMSE%) = 15.97%). This study demonstrated the benefits of UAV imagery-based multi-feature fusion using machine learning for above-ground biomass estimation in oat breeding nurseries, holding promise for enhancing the efficiency of oat breeding through UAV-based phenotyping and crop management practices.

show abstract

Section: Resultsmentioning

confidence: 99%

Utilizing Spectral, Structural and Textural Features for Estimating Oat Above-Ground Biomass Using UAV-Based Multispectral Data and Machine Learning

Dhakal,

Maimaitijiang,

Chang

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…These data could be used to develop calibrations between field measurements and UAV outputs. One part of the data (the tree from the landscape data set) has already been used to develop calibrations (Bossoukpe, Faye, et al, 2021; Bossoukpe, Ndiaye, et al, 2021; Taugourdeau et al, 2022). One possible interest could be to carry out a meta‐analysis to develop calibrations on a larger scale across different types of vegetation, but also between different UAVs and flight plans.…”

Section: Potential Uses Of Theses Data Setsmentioning

confidence: 99%

Unmanned aerial vehicle outputs and associated field measurements of the herbaceous and tree layers of the Senegalese savannah

et al. 2023

Self Cite

View full text Add to dashboard Cite

Unmanned aerial vehicles (UAVs) are widely used to assess the vegetation of an ecosystem. This report presents three data sets from UAVs used on a savannah-type ecosystem in Senegal. Two data sets contain an orthomosaic and a surface and terrain digital model (UAV outputs), along with associated field measurements on the tree and herbaceous layers, while the third data set contains an orthomosaic and digital surface model, along with field measurements on the herbaceous layers. One data set was compiled over the entire rainy season on the same plot (plot dataset). One was compiled across a landscape at the 7000 ha Dahra Djoloff research center (CRZdahra Centre de recherche zootechnique de Dahra). The third data set was compiled across Senegal from the North to the Southeast of the country with wide variability in climate and soil conditions. All the data sets are in Open Access in the Zenodo repository. These data sets could be used in many different studies: calibration between field measurements and UAV outputs on a large scale, or used as an intermediate step between field and satellite images.

show abstract

“…Schulze-Bruninghoff et al ( 2021) suggested a method combining lidar-derived metrics (sum of voxels, canopy height model, and canopy surface structure) and vegetation spectral properties to estimate fresh and dry biomass in a machine learning approach. In addition, by combining ground camera and UAS, Taugourdeau et al (2022) extracted vegetation metrics from RGB bands and height to estimate aboveground biomass in a random forest model. Therefore, the uncertainty of biomass estimation is limited to the uncertainty in CHM, data resolution, and the type of land treatments (grazed or ungrazed).…”

Section: Introductionmentioning

confidence: 99%

Voxel Volumes and Biomass: estimating vegetation volume and litter accumulation of exotic annual grasses using automated ultra-high resolution SfM and advanced classification techniques

Enterkine,

Hojatimalekshah,

Vermillion

et al. 2023

Preprint

View full text Add to dashboard Cite

Rangelands and semi-arid ecosystems are subject to increasing changes in ecologic makeup from a collection of factors. In much of the northern Great Basin, rangelands invaded by exotic annual grasses such as cheatgrass (Broumus tectorum) and medusahead (Taeniatherum caput-medusae) are experiencing an increasingly short fire cycle which is compounding and persistent. Improving and expanding ground-based field methods for measuring above-ground biomass (AGB) may enable more sample collections across a landscape and over succession regimes, and better harmonize with other remote sensing techniques. Developments and increased adoption of uncrewed aerial vehicles and instrumentation for vegetation monitoring are enabling greater understanding of vegetation in many ecosystems. Research towards understanding the relationship of traditional field measurements with newer aerial platforms in rangeland environments is growing rapidly, and there is increasing interest in exploring the potential use both to quantify AGB and fine fuel load at pasture scales. Our study here uses relatively inexpensive handheld photography with custom sampling frames to collect and automatically reconstruct 3D-models of the vegetation within 0.2 m2 quatrats (n = 288). Next, we examine the relationship between volumetric estimates of vegetation to compare with biomass. We found that volumes calculated with 0.5 cm voxel sizes (0.125 cm3) most closely represented the range of biomass weights. We further develop methods to classify ground points, finding a 2% reduction in predictive ability compared to using the true ground surface. Overall, our reconstruction workflow had an R2 of 0.42, further emphasizing the importance of high-resolution imagery and reconstruction techniques. Ultimately, we conclude that more work is needed of increasing extents (such as from UAS) to better understand and constrain uncertainties in volumetric estimations of biomass in ecosystems with high amounts of invasive annual grasses and fine fuel litter.

show abstract

Estimating herbaceous aboveground biomass in Sahelian rangelands using Structure from Motion data collected on the ground and by UAV

Abstract: This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Cited by 11 publications

References 45 publications

Utilizing Spectral, Structural and Textural Features for Estimating Oat Above-Ground Biomass Using UAV-Based Multispectral Data and Machine Learning

Utilizing Spectral, Structural and Textural Features for Estimating Oat Above-Ground Biomass Using UAV-Based Multispectral Data and Machine Learning

Unmanned aerial vehicle outputs and associated field measurements of the herbaceous and tree layers of the Senegalese savannah

Voxel Volumes and Biomass: estimating vegetation volume and litter accumulation of exotic annual grasses using automated ultra-high resolution SfM and advanced classification techniques

Contact Info

Product

Resources

About