Search

Global Ecosystem Dynamics Investigation (GEDI) is a lidar system on-board the International Space Station designed to study forest ecosystems. However, GEDI footprint low accuracy geolocation is a major impediment to the optimal benefit of the data. We thus proposed a geolocation correction method, GeoGEDI, only based on high-resolution digital elevation models (DEMs) and GEDI derived ground elevations. For each footprint, an error map between GEDI ground estimates and reference DEM was computed, and a flow accumulation algorithm was used to retrieve the optimal footprint position. GeoGEDI was tested on 150 000 footprints in Landes and Vosges, two French forests with various stands and topographic conditions. It was applied to GEDI versions 1 (v1) and 2 (v2), by either a single or four full-power laser beam tracks. GeoGEDI output accuracy was evaluated by analyzing shift distributions and comparing GEDI ground elevations and surface heights to reference data. GeoGEDI corrections were greater for v1 than for v2 and agreed with errors published by NASA. Within forests, GeoGEDI improved the root mean square error (RMSE) of ground elevation in Landes by 26.8% (0.34 m) and by 13.3% (0.14 m) for v1 and v2, respectively. For Vosges, ground elevation RMSE improved by 59.6% (3.82 m) and 36.2% (1.41 m), for v1 and v2, respectively. Regarding surface heights, except for v2 in Landes, where insufficient variations in topography combined to GEDI ground detection issues might have penalized the adjustment, GeoGEDI improved GEDI estimates. Using GeoGEDI showed efficient to improve positioning bias and precision.

The strong societal demand to reduce pesticide use and adaptation to climate change challenges the capacities of phenotyping new varieties in the vineyard. High-throughput phenotyping is a way to obtain meaningful and reliable information on hundreds of genotypes in a limited period. We evaluated traits related to growth in 209 genotypes from an interspecific grapevine biparental cross, between IJ119, a local genitor, and Divona, both in summer and in winter, using several methods: fresh pruning wood weight, exposed leaf area calculated from digital images, leaf chlorophyll concentration, and LiDAR-derived apparent volumes. Using high-density genetic information obtained by the genotyping by sequencing technology (GBS), we detected 6 regions of the grapevine genome [quantitative trait loci (QTL)] associated with the variations of the traits in the progeny. The detection of statistically significant QTLs, as well as correlations (R2) with traditional methods above 0.46, shows that LiDAR technology is effective in characterizing the growth features of the grapevine. Heritabilities calculated with LiDAR-derived total canopy and pruning wood volumes were high, above 0.66, and stable between growing seasons. These variables provided genetic models explaining up to 47% of the phenotypic variance, which were better than models obtained with the exposed leaf area estimated from images and the destructive pruning weight measurements. Our results highlight the relevance of LiDAR-derived traits for characterizing genetically induced differences in grapevine growth and open new perspectives for high-throughput phenotyping of grapevines in the vineyard.

The spatial distribution of Leaf Area Density (LAD) in a tree canopy has fundamental functions in ecosystems. It can be measured through a variety of methods, including voxel-based methods applied to LiDAR point clouds. A theoretical study recently compared the numerical errors of these methods and showed that the bias-corrected Maximum Likelihood Estimator was the most efficient. However, it ignored (i) wood volumes, (ii) vegetation sub-grid clumping, (iii) the instrument effective footprint, and (iv) was limited to a single viewpoint. In practice, retrieving LAD is not straightforward, because vegetation is not randomly distributed in sub-grids, beams are divergent, and forestry plots are sampled from more than one viewpoint to mitigate occlusion. In the present article, we extend the previous formulation to (i) account for both wood volumes and hits, (ii) rigorously include correction terms for vegetation and instrument characteristics, and (iii) integrate multiview data. Two numerical experiments showed that the new approach entailed reduction of bias and errors, especially in the presence of wood volumes or when multiview data are available for poorly-explored volumes. In addition to its conciseness, completeness, and efficiency, this new formulation can be applied to multiview TLS—and also potentially to UAV LiDAR scanning—to reduce errors in LAD estimation.

Reliable measurements of the 3D distribution of Leaf Area Density (LAD) in forest canopy are crucial for describing and modelling microclimatic and eco-physiological processes involved in forest ecosystems functioning. To overcome the obvious limitations of direct measurements, several indirect methods have been developed, including methods based on Terrestrial LiDAR scanning (TLS). This work focused on various LAD estimators used in voxel-based approaches. LAD estimates were compared to reference measurements at branch scale in laboratory, which offered the opportunity to investigate in controlled conditions the sensitivity of estimations to various factors such as voxel size, distance to scanner, leaf morphology (species), type of scanner and type of estimator. We found that all approaches to retrieve LAD estimates were highly sensitive to voxel size whatever the species or scanner and to distance to the FARO scanner. We provided evidence that these biases were caused by vegetation heterogeneity and variations in the effective footprint of the scanner. We were able to identify calibration functions that could be readily applied when vegetation and scanner are similar to those of the present study. For different vegetation and scanner, we recommend replicating our method, which can be applied at reasonable cost. While acknowledging that the test conditions in the laboratory were very different from those of the measurements taken in the forest (especially in terms of occlusion), this study revealed existence of strong biases, including spatial biases. Because the distance between scanner and vegetation varies in field scanning, these biases should occur in a similar manner in the field and should be accounted for in voxel-based methods but also in gap-fraction methods.