Monitoring changes of forest height in California

The impact of geolocation uncertainty on GEDI tropical forest canopy height estimation and change monitoring

Forests are one of the key elements in ecological transition policies in Europe. Sustainable forest management is needed in order to optimise wood harvesting, while preserving carbon storage, biodiversity and other ecological functions. Forest managers and public bodies need improved and cost-effective forest monitoring tools. Research studies have been carried out to assess the use of optical and radar images for producing forest height or biomass maps. The main limitations are the quantity, quality and representativeness of the reference data for model training. The Global Ecosystem Dynamics Investigation (GEDI) mission (full waveform LiDAR on board the International Space Station) has provided an unprecedented number of forest canopy height samples from 2019. These samples could be used to improve reference datasets. This paper aims to present and validate a method for estimating forest dominant height from open access optical and radar satellite images (Sentinel-1, Sentinel-2 and ALOS-2 PALSAR-2), and then to assess the use of GEDI samples to replace field height measurements in model calibration. Our approach combines satellite image features and dominant height measurements, or GEDI metrics, in a Support Vector Machine regression algorithm, with a feature selection process. The method is tested on mixed uneven-aged broadleaved and coniferous forests in France. Using dominant height measurements for model training, the cross-validation shows 7.3 to 11.6% relative Root Mean Square Error (RMSE) depending on the forest class. When using GEDI height metrics instead of field measurements for model training, errors increase to 12.8–16.7% relative RMSE. This level of error remains satisfactory; the use of GEDI could allow the production of dominant height maps on large areas with better sample representativeness. Future work will focus on confirming these results on new study sites, improving the filtering and processing of GEDI data, and producing height maps at regional or national scale. The resulting maps will help forest managers and public bodies to optimise forest resource inventories, as well as allow scientists to integrate these cartographic data into climate models.

Direct use of large-footprint lidar waveforms to estimate aboveground biomass

Identification of forest fuels is a key step for forest fire prevention since they provide valuable information of fire behavior. This study assesses NASA’s Global Ecosystem Dynamics Investigation (GEDI) system to classify fuel types in Mediterranean environments according to the Prometheus model in a forested area of NE Spain. We used 59,554 GEDI footprints and extracted variables related to height metrics, canopy profile metrics, and aboveground biomass density estimates from products L2A, L2B, and L4A, respectively. Four quality filters were applied to discard high uncertainty data, reducing the initial footprints to 9,703. Spectral indices from Landsat-8 OLI scenes were created to test the effect of their integration with GEDI variables on fuel types estimation. Ground-truth data were comprised of Prometheus fuel types estimated in two previous studies. Only the types that matched in each GEDI footprint in both studies were used, resulting in a final sample of 1,112 footprints. Spearman’s correlation coefficient, Kruskal-Wallis and Dunn’s tests determined the variables to be included in the classification models: the relative height at the 85th percentile, the Plant Area Index, and the Aboveground Biomass Density from GEDI, and the brightness from Landsat-8 OLI. Best performances were achieved with Random Forest (RF) and Support Vector Machine with radial kernel (SVM-R), which were lower including only GEDI variables (accuracies: RF and SVM-R = 61.54 %) than integrating the brightness from Landsat-8 OLI (accuracies: RF = 83.71 %, SVM-R = 81.90 %). These results allow validating GEDI for fuel type classification of Prometheus model, constituting a promising information for forest management over large areas.

Continuous large-scale mapping of forest canopy height is crucial for estimating and reporting forest carbon content, analyzing forest degradation and restoration, or to model ecosystem variables such as aboveground biomass. Over the last years, the spaceborne Light Detection and Ranging (LiDAR) sensor specifically designed to acquire forest structure information, Global Ecosystem Dynamics Investigation (GEDI), has been used to extract forest canopy height information over large areas. Yet, GEDI has no spatial coverage for most forested areas in Canada and other high latitude regions. On the other hand, the spaceborne LiDAR called Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) provides a global coverage but was not specially developed to study forested ecosystems. Nonetheless, both spaceborne LiDAR sensors obtain point-based information, making spatially continuous forest canopy height estimation very challenging. This study compared the performance of both spaceborne LiDAR, GEDI and ICESat-2, combined with ALOS-2/PALSAR-2 and Sentinel-1 and -2 data to produce continuous canopy height maps in Canada for the year 2020. A set-aside dataset and airborne LiDAR (ALS) from a national LiDAR campaign were used for accuracy assessment. Both maps overestimated canopy height in relation to ALS data, but GEDI had a better performance than ICESat-2 with a mean difference (MD) of 0.9 m and 2.9 m, and a root mean square error (RMSE) of 4.2 m and 5.2 m, respectively. However, as both GEDI and ALS have no coverage in most of the hemi-boreal forests, ICESat-2 captures the tall canopy heights expected for these forests better than GEDI. PALSAR-2 HV polarization was the most important covariate to predict canopy height, showing the great potential of L-band in comparison to C-band from Sentinel-1 or optical data from Sentinel-2. The approach proposed here can be used operationally to produce annual canopy height maps for areas that lack GEDI and ICESat-2 coverage.

Modelling forest volume with small area estimation of forest inventory using GEDI footprints as auxiliary information

The Global Ecosystem Dynamics Investigation (GEDI), a new spaceborne LiDAR system of the National Aeronautics and Space Administration (NASA), has the potential to revolutionize global measurements of vertical vegetation structure. However, GEDI performance among different forest types and factors influencing GEDI performance needs to be evaluated against similar measurements from existing airborne LiDAR platforms. Ideally, comparisons across diverse forest types will inform future work quantifying biomass or mapping species habitats. Thus, we compared the second version of GEDI L2A product (GEDI V2) with Airborne Observation Platform (AOP) leaf-on LiDAR data across 33 National Ecological Observation Network (NEON) sites. Comparisons were made for ground elevation and relative height (RH) of GEDI with simulated airborne laser scanning (ALS) waveforms from discrete point cloud LiDAR. Results indicated that GEDI V2 obtained high accuracy on ground elevation and RH100 estimations (3σ) with RMSEs of 1.38 m and 2.62 m, respectively. GEDI produced forest height estimations (RH100) for all 12 forest types with a %RMSE below 25%. GEDI RHs were sensitive to ground finding accuracy, and GEDI performance of RH estimation varied from forest profiles of different forest types. For factors influencing GEDI performance, greater than 21% of GEDI RH95 and 33% of ground elevation variations can be explained by land surface attributes, observing sensor system characteristics, and the collection time differences between GEDI and NEON LiDAR. Furthermore, geolocation error remains an essential factor affecting GEDI performance, which varies among forest and land cover types, especially for canopy height estimation. The findings reported here can provide insights to guide and enhance future GEDI-based global forest structure mapping and applications.

The development of high resolution mapping models of forest attributes based on employing machine or deep learning techniques has increasingly accelerated in the last couple of years. The consequence of this is the widespread availability of multiple sources of information, which can either lead to a potential confusion, or to a possibility to get an "extended” insight into the state of our forests by interpreting these sources jointly. This contribution aims at addressing the latter, by relying on the Bayesian model averaging (BMA) approach.BMA is a method that can be used in building a consensus from an ensemble of different model predictions. It can be seen as weighted mean of different predictions with weights reflecting the predictive performances of different models, or as a finite mixture model which estimates the probability that each observation from the independent validation dataset has been generated by one of the models belonging to the ensemble. BMA can thus be used to diagnose and understand the difference in the predictions and to possibly interpret them.The predictions in our case are the forest canopy height estimations for the metropolitan France coming from 5 different AI models [1-5], while the independent validation dataset comes from the French National Forest Inventory (NFI) disposing with some 6000 plots per year, distributed across the territory of interest. For every plot we have several measurements/estimations of the forest canopy height out of which the following two are considered in this study: h_m – the maximum total height (from the tree's base level to the terminal bud of the tree's main stem) measured within the plot, and h_dom – the average height of the seven largest dominant trees per hectare.In this contribution we present for every considered plot the dominant model with respect to both references i.e. the model having the highest probability to be the one generating measurements/estimations at NFI plot (h_m and h_dom). We present as well as the respective inter-model and the intra-model variance estimations, allowing us to propose a series of hypotheses concerning the established differences between predictions of individual models in function of their specificities.[1] Schwartz, M., et al.: FORMS: Forest Multiple Source height, wood volume, and biomass maps in France at 10 to 30 m resolution based on Sentinel-1, Sentinel-2, and Global Ecosystem Dynamics Investigation (GEDI) data with a deep learning approach, Earth Syst. Sci. Data, 15, 4927–4945, 2023, https://doi.org/10.5194/essd-15-4927-2023[2] Lang, N., et al.: A high-resolution canopy height model of the Earth, Nat Ecol Evol 7, 1778–1789, 2023. https://doi.org/10.1038/s41559-023-02206-6[3] Morin, D. et al.: Improving Heterogeneous Forest Height Maps by Integrating GEDI-Based Forest Height Information in a Multi-Sensor Mapping Process, Remote Sens., 14, 2079. 2022, https://doi.org/10.3390/rs14092079[4] Potapov, P., et al.: Mapping global forest canopy height through integration of GEDI and Landsat data, Remote Sensing of Environment, 253, 2021, https://doi.org/10.1016/j.rse.2020.112165.[5] Liu, S. et al.: The overlooked contribution of trees outside forests to tree cover and woody biomass across Europe, Sci. Adv. 9, eadh4097, 2023, 10.1126/sciadv.adh4097.

Spatially-explicit information on canopy fuel parameters is key for wildfire propagation modelling, emission estimations and risk assessment. This work aims to develop easily-replicable methods to estimate critical fuel canopy parameters from spaceborne LiDAR observations acquired by the Global Ecosystem Dynamics Investigation (GEDI) sensor onboard the International Space Station. GEDI-like pseudowaveforms were modelled from discrete Airborne Laser Scanning (ALS) data and used to select the best GEDI predictor metrics to derive European wall-to-wall forest height and canopy cover maps. Then, GEDI spaceborne footprints were used to generate continental maps of canopy parameters through a two-steps approach: 1) Spatial interpolation of GEDI footprints inside homogeneous forest fuel type polygons, and 2) Modelling machine learning algorithms for the forest fuel type polygons without GEDI footprints inside, using auxiliary multispectral and RADAR imagery and biophysical variables. Our results show the capabilities of remote sensing and GEDI to estimate and map the spatial patterns of critical forest canopy fuel parameters in fire risk prevention and contribute to generating the necessary tools to develop an integrated risk-wise strategy that reduces fire vulnerability of ecosystems across Europe.

Monitoring forest structural properties is critical for a range of applications because structure is key to understanding and quantifying forest biophysical functioning, including stand dynamics, evapotranspiration, habitat, and recovery from disturbances. Monitoring of forest structural properties at desirable frequencies and cost globally is enabled by space-borne LiDAR missions such as the global ecosystem dynamics investigation (GEDI) mission. This study assessed the accuracy of GEDI estimates for canopy height, total plant area index (PAI), and vertical profile of plant area volume density (PAVD) and elevation over a gradient of canopy height and terrain slope, compared to estimates derived from airborne laser scanning (ALS) across two forest age-classes in the Central Highlands region of south-eastern Australia. ALS was used as a reference dataset for validation of GEDI (Version 2) dataset. Canopy height and total PAI analyses were carried out at the landscape level to understand the influence of beam-type, height of the canopy, and terrain slope. An assessment of GEDI’s terrain elevation accuracy was also carried out at the landscape level. The PAVD profile evaluation was carried out using footprints grouped into two forest age-classes, based on the areas of mountain ash (Eucalyptus regnans) forest burnt in the Central Highlands during the 1939 and 2009 wildfires. The results indicate that although GEDI is found to significantly under-estimate the total PAI and slightly over-estimate the canopy height, the GEDI estimates of canopy height and the vertical PAVD profile (above 25 m) show a good level of accuracy. Both beam-types had comparable accuracies, with increasing slope having a slightly detrimental effect on accuracy. The elevation accuracy of GEDI found the RMSE to be 10.58 m and bias to be 1.28 m, with an R2 of 1.00. The results showed GEDI is suitable for canopy densities and height in complex forests of south-eastern Australia.

Bridging spatio-temporal gaps in ALS data using Landsat time series and forest disturbance-recovery metrics via multi-task neural networks

Quantifying forest biomass carbon (C) stock is critical for determining the regional carbon balance, but, a lack of both field observations spanning large climatic gradients and proper upscaling methods which take the spatial pattern into account, means there is little knowledge regarding forest C stock at high spatial resolutions. Here, we address this problem by combining a deep convolutional neural network (CNN) algorithm with in situ measurements, Global Ecosystem Dynamics Investigation (GEDI) observations, and Landsat and PALSAR-2 images to develop a new, spatially explicit estimate of forest aboveground carbon density (ACD) circa 2020 at a 30 m spatial resolution for northeast China, home to nearly one-third of China’s forested area. The result yields a high coefficient of determination (R2) of 0.83 and a relatively low root mean squared error (RMSE) of 5.28 MgC ha-1, and is superior to traditional pixel-based and in situ based methods. Through linking in situ measurements with nearly 0.13 million GEDI observations, we obtained important samples across spatially variable environmental conditions, and in remote and rugged regions (when increasing the number of GEDI samples, RMSE decreased by 73.5%). CNN was able to extract important spatial patterns and performed well in capturing the spatial variation of forest carbon density. We also propose a CNN-based perturbation method to rank variable importance, which shows that the distribution of forest C storage is mainly determined by precipitation and forest age. Based on the proposed method, the local forest aboveground biomass C stock is estimated to be 3.52 ± 0.10 PgC, with an age-related forest aboveground biomass C sink of 7.94 TgC year-1 before 2060. Terrestrial ecosystem models generally underestimate the regional C stock, partially because of biases in forest age simulations. The study highlights the importance of using deep learning methods to gain further process understanding of forest carbon dynamics under climate change.

More of the Amazon rainforest is disturbed each year than completely deforested, but the impact of these disturbances on the carbon cycle remains poorly understood. Recent algorithmic advances using optical and radar remote sensing have improved detection of disturbances at fine spatiotemporal resolution, but quantifying changes in forest structure and biomass associated with these detected disturbances has proven challenging. The Global Ecosystem Dynamics Investigation (GEDI), a spaceborne lidar mission collecting data from 2019 to 2023, provides an opportunity to address this problem. GEDI captured billions of measurements of forest height, plant area, and understory structure within ∼25-m diameter footprints scattered across the tropics. Though the instrument had no guaranteed repeat cycle, it sometimes sampled nearby locations twice; some of these spatially near-coincident footprints happened to measure forest structure before and after a detected disturbance, providing information on how disturbance affected aboveground biomass. In this study, we developed an efficient general-purpose open-source pipeline for identifying spatially coincident footprints, which is a computationally complex task, and used the pipeline to find over 13,700 footprint pairs with intervening disturbance events across the Amazon biome. We also identified a set of ∼65,000 spatially near-coincident footprint pairs that lacked an intervening disturbance but came from regions with similar disturbance threat; these provided a control dataset to evaluate the effectiveness of estimating forest structure and biomass changes with nearby footprints. Analysis of this Amazon-wide dataset demonstrated that GEDI was able to measure statistically significant canopy height and biomass deficits following non-stand-replacing disturbances as small as 900 m2 (30 m × 30 m). GEDI's unique three-dimensional view of forest structure also reflected the effects of different intensities of fire disturbance, including 20% of burned areas where the upper canopy retained most of its height, but the understory suffered substantial foliage losses. Finally, we modeled the relationship between Landsat and Sentinel-1 disturbance detection parameters and GEDI-estimated percent biomass loss, showing that certain satellite-derived intensity metrics are correlated with increasing biomass loss and identifying temporal trends in biomass loss and recovery following disturbance. This work represents an important step towards the development of a pan-tropical, spatially explicit system for tracking carbon losses and structural changes arising from forest disturbance.

Cross-modal fusion approach with multispectral, LiDAR, and SAR data for forest canopy height mapping in mountainous region

Understanding the resilience of tropical vegetation, its ability to recover from disturbance, is fundamental to assess future responses to environmental and climatic fluctuations. The Amazon rainforest has been identified as a potential tipping element in the Earth’s climate system and there is mounting concern over its persistent degradation. Extreme climate events and continued logging, forest fire and fragmentation threaten the Amazon’s structural integrity and its role as a carbon sink, with remotely sensed data providing observational evidence of resilience loss since the early 2000s. Fragmentation and degradation of tropical forest is suggested to slow recovery from perturbations, ensuing a potential to destabilise the rainforest and cause widespread transition from forest to savanna-like ecosystem state.Remotely sensed LiDAR data provides a structural blueprint of forest canopy. The Global Ecosystem Dynamics Investigation (GEDI) spaceborne LiDAR characterises a new era of large-scale forest height quantification, with capabilities to further understand forest structure, and therefore forest response to perturbation across the entire Amazon. Although GEDI’s capabilities have been realised in boreal forest early disturbance monitoring, and to assess growth rates of tropical secondary forest, research thus far is yet to assess its ability to identify tropical forest of various degradation and recovery including logged, burned and fragmented over increasing timescales of recovery. Forest degraded by burning is characterised by different structure than selectively logged, or edge forest, and validating the ability of GEDI to represent these states is essential for identifying alternative forest states.  Here, we investigate the potential of the GEDI LiDAR mission to map tropical forest along a gradient of degradation to recovery. A combination of ground data, MapBiomas secondary forest and burned area products are utilised to classify perturbed forest. We then assess the correspondence of GEDI waveform metrics including relative height and canopy cover, extracted from 2A and 2B products using the newly developed R package ‘chewie’, with airborne LiDAR across the Brazilian Amazon. This research will inform further tropical forest alternative-state study, whilst the assessment of GEDI’s structural capability to represent degraded forest types provides valuable information for forest restoration status to support post-degradation management strategies.