Current and planned space missions will produce aboveground biomass density data products at varying spatial resolution. Calibration and validation of these data products is critically dependent on the existence of field estimates of aboveground biomass and coincident remote sensing data from airborne or terrestrial lidar. There are few places that meet these requirements, and they are mostly in the northern hemisphere and temperate zone. Here we summarize the potential for low-altitude drones to produce new observations in support of mission science. We describe technical requirements for producing high-quality measurements from autonomous platforms and highlight differences among commercially available laser scanners and drone aircraft. We then describe a case study using a heavy-lift autonomous helicopter in a temperate mountain forest in the southern Czech Republic in support of calibration and validation activities for the NASA Global Ecosystem Dynamics Investigation. Low-altitude flight using drones enables the collection of ultra-high-density point clouds using wider laser scan angles than have been possible from traditional airborne platforms. These measurements can be precise and accurate and can achieve measurement densities of thousands of points · m⁻². Analysis of surface elevation measurements on a heterogeneous target observed 51 days apart indicates that the realized range accuracy is 2.4 cm. The single-date precision is 2.1–4.5 cm. These estimates are net of all processing artifacts and geolocation errors under fully autonomous flight. The 3D model produced by these data can clearly resolve branch and stem structure that is comparable to terrestrial laser scans and can be acquired rapidly over large landscapes at a fraction of the cost of traditional airborne laser scanning.

NASA’s Global Ecosystem Dynamics Investigation (GEDI) mission will collect waveform lidar data at a dense sample of ∼25 m footprints along ground tracks paralleling the orbit of the International Space Station (ISS). GEDI’s primary science deliverable will be a 1 km grid of estimated mean aboveground biomass density (Mg ha⁻¹), covering the latitudes overflown by ISS (51.6 °S to 51.6 °N). One option for using the sample of waveforms contained within an individual grid cell to produce an estimate for that cell is hybrid inference, which explicitly incorporates both sampling design and model parameter covariance into estimates of variance around the population mean. We explored statistical properties of hybrid estimators applied in the context of GEDI, using simulations calibrated with lidar and field data from six diverse sites across the United States. We found hybrid estimators of mean biomass to be unbiased and the corresponding estimators of variance appeared to be asymptotically unbiased, with under-estimation of variance by approximately 20% when data from only two clusters (footprint tracks) were available. In our study areas, sampling error contributed more to overall estimates of variance than variability due to the model, and it was the design-based component of the variance that was the source of the variance estimator bias at small sample sizes. These results highlight the importance of maximizing GEDI’s sample size in making precise biomass estimates. Given a set of assumptions discussed here, hybrid inference provides a viable framework for estimating biomass at the scale of a 1 km grid cell while formally accounting for both variability due to the model and sampling error.

The Global Ecosystem Dynamics Investigation (GEDI) lidar is a multibeam laser altimeter on the International Space Station (ISS). GEDI is the first spaceborne instrument designed to measure vegetation height and to quantify aboveground carbon stocks in temperate and tropical forests and woodlands. This document describes the algorithm theoretical basis underpinning the development of the GEDI Level‐4A (GEDI04_A) footprint aboveground biomass density (AGBD) data product. The GEDI04_A data product contains estimates of AGBD for individual GEDI footprints and associated prediction intervals. The algorithm uses GEDI02_A relative height metrics and 13 linear models to predict AGBD in 32 combinations of plant functional type and world region within the observation limits of the ISS. GEDI04_A models for the release 1 and release 2 data products were developed using 8,587 quality‐filtered simulated GEDI waveforms associated with field estimates of AGBD in 21 countries. Although this is the most geographically comprehensive data available for the development of AGBD models using lidar remote sensing, important regions are underrepresented, including the forests of continental Asia, deciduous broadleaf forests and savannas of the dry tropics, and evergreen broadleaf forests north of Australia. We describe the scientific and statistical assumptions required to develop globally representative estimates of AGBD using GEDI lidar, including generalization beyond training data, and exclusion of GEDI02_A observations that do not meet requirements of the GEDI04_A algorithm. The footprint‐level predictions generated by this process provide globally comprehensive estimates of AGBD. These footprint‐level predictions are a prerequisite for the GEDI04_B gridded AGBD data product.

We applied a supervised individual-tree segmentation algorithm to ultra-high-density drone lidar in a temperate mountain forest in the southern Czech Republic. We compared the number of trees correctly segmented, stem diameter at breast height (DBH), and tree height from drone-lidar segmentations to field-inventory measurements and segmentations from terrestrial laser scanning (TLS) data acquired within two days of the drone-lidar acquisition. Our analysis detected 51% of the stems >15 cm DBH, and 87% of stems >50 cm DBH. Errors of omission were much more common for smaller trees than for larger ones, and were caused by removal of points prior to segmentation using a low-intensity and morphological filter. Analysis of segmented trees indicates a strong linear relationship between DBH from drone-lidar segmentations and TLS data. The slope of this relationship is 0.93, the intercept is 4.28 cm, and the r² is 0.98. However, drone lidar and TLS segmentations overestimated DBH for the smallest trees and underestimated DBH for the largest trees in comparison to field data. We evaluate the impact of random error in point locations and variation in footprint size, and demonstrate that random error in point locations is likely to cause an overestimation bias for small-DBH trees. A Random Forest classifier correctly identified broadleaf and needleleaf trees using stem and crown geometric properties with overall accuracy of 85.9%. We used these classifications and DBH estimates from drone-lidar segmentations to apply allometric scaling equations to segmented individual trees. The stand-level aboveground biomass (AGB) estimate using these data is 76% of the value obtained using a traditional field inventory. We demonstrate that 71% of the omitted AGB is due to segmentation errors of omission, and the remaining 29% is due to DBH estimation errors. Our analysis indicates that high-density measurements from low-altitude drone flight can produce DBH estimates for individual trees that are comparable to TLS. These data can be collected rapidly throughout areas large enough to produce landscape-scale estimates. With additional refinement, these estimates could augment or replace manual field inventories, and could support the calibration and validation of current and forthcoming space missions.

Click on the article title to read more.