Search

Satellite radar backscatter contains unique information on land surface moisture, vegetation features, and surface roughness and has thus been used in a range of Earth science disciplines. However, there is no single global radar data set that has a relatively long wavelength and a decades-long time span. We here provide the first long-term (since 1992), high-resolution (∼8.9 km instead of the commonly used ∼25 km resolution) monthly satellite radar backscatter data set over global land areas, called the long-term, high-resolution scatterometer (LHScat) data set, by fusing signals from the European Remote Sensing satellite (ERS; 1992–2001; C-band; 5.3 GHz), Quick Scatterometer (QSCAT, 1999–2009; Ku-band; 13.4 GHz), and the Advanced SCATterometer (ASCAT; since 2007; C-band; 5.255 GHz). The 6-year data gap between C-band ERS and ASCAT was filled by modelling a substitute C-band signal during 1999–2009 from Ku-band QSCAT signals and climatic information. To this end, we first rescaled the signals from different sensors, pixel by pixel. We then corrected the monthly signal differences between the C-band and the scaled Ku-band signals by modelling the signal differences from climatic variables (i.e. monthly precipitation, skin temperature, and snow depth) using decision tree regression. The quality of the merged radar signal was assessed by computing the Pearson r, root mean square error (RMSE), and relative RMSE (rRMSE) between the C-band and the corrected Ku-band signals in the overlapping years (1999–2001 and 2007–2009). We obtained high Pearson r values and low RMSE values at both the regional (r≥0.92, RMSE ≤ 0.11 dB, and rRMSE ≤ 0.38) and pixel levels (median r across pixels ≥ 0.64, median RMSE ≤ 0.34 dB, and median rRMSE ≤ 0.88), suggesting high accuracy for the data-merging procedure. The merged radar signals were then validated against the European Space Agency (ESA) ERS-2 data, which provide observations for a subset of global pixels until 2011, even after the failure of on-board gyroscopes in 2001. We found highly concordant monthly dynamics between the merged radar signals and the ESA ERS-2 signals, with regional Pearson r values ranging from 0.79 to 0.98. These results showed that our merged radar data have a consistent C-band signal dynamic. The LHScat data set (https://doi.org/10.6084/m9.figshare.20407857; Tao et al., 2023) is expected to advance our understanding of the long-term changes in, e.g., global vegetation and soil moisture with a high spatial resolution. The data set will be updated on a regular basis to include the latest images acquired by ASCAT and to include even higher spatial and temporal resolutions.

International audience; Intact tropical rainforests have been exposed to severe droughts in recent decades, which may threaten their integrity, their ability to sequester carbon, and their capacity to provide shelter for biodiversity. However, their response to droughts remains uncertain due to limited high-quality, long-term observations covering extensive areas. Here, we examined how the upper canopy of intact tropical rainforests has responded to drought events globally and during the past 3 decades. By developing a long pantropical time series (1992 to 2018) of monthly radar satellite observations, we show that repeated droughts caused a sustained decline in radar signal in 93%, 84%, and 88% of intact tropical rainforests in the Americas, Africa, and Asia, respectively. Sudden decreases in radar signal were detected around the 1997–1998, 2005, 2010, and 2015 droughts in tropical Americas; 1999–2000, 2004–2005, 2010–2011, and 2015 droughts in tropical Africa; and 1997–1998, 2006, and 2015 droughts in tropical Asia. Rainforests showed similar low resistance (the ability to maintain predrought condition when drought occurs) to severe droughts across continents, but American rainforests consistently showed the lowest resilience (the ability to return to predrought condition after the drought event). Moreover, while the resistance of intact tropical rainforests to drought is decreasing, albeit weakly in tropical Africa and Asia, forest resilience has not increased significantly. Our results therefore suggest the capacity of intact rainforests to withstand future droughts is limited. This has negative implications for climate change mitigation through forest-based climate solutions and the associated pledges made by countries under the Paris Agreement.

Satellite radar backscatter contains unique information on land surface moisture, vegetation features, and surface roughness, and can be acquired in all weather conditions, thus has been used in a range of earth science disciplines. However, there is no single global radar data set that spans more than two decades. This has limited the use of radar data for trend analysis over extended time intervals. We here provide the first long-term (since 1992), high resolution (~8.9 km) satellite radar backscatter data set over global land areas, the C-band Scatterometer (CScat) data set, by fusing signals from European Remote Sensing satellite (ERS, 1992–2001, C-band, 5.3 GHz), Quick Scatterometer (QSCAT, 1999–2009, Ku-band, 13.4 GHz), and the Advanced Scatterometer (ASCAT, since 2007, C-band, 5.255 GHz). The six-year data gap between C-band ERS and ASCAT was filled out by modelling an equivalent C-band signal during 1999–2009 from Ku-band QSCAT signals and climatic information. Towards this purpose, we first rescaled the signals from different sensors, pixel by pixel, using a new signal rescaling method that is robust to limited overlapping observations among sensors. We then corrected the monthly signal differences between the C-band and the scaled Ku-band signals, by modelling the signal differences from climatic variables (i.e., monthly precipitation, skin temperature, and snow depth) using decision tree regression. The quality of the merged radar signal was assessed by computing the Pearson r, Root Mean Square Error (RMSE), and relative RMSE (rRMSE) between the C-band and the corrected Ku-band signals in the overlapping years (1999–2001 and 2007–2009). We obtained high Pearson r values and low RMSE values at both the regional (r ≥ 0.93, RMSE ≤ 0.16, rRMSE ≤0.37) and pixel levels (median r across pixels ≥ 0.80, median RMSE ≤ 0.38, median rRMSE ≤ 0.64), suggesting high accuracy for the data merging procedure. The merged radar signal was then validated with a continuous ERS-2 data set available between 1995 and 2011. ERS-2 stopped working in full mode after 2001 but observations are occasionally available for a subset of the pixels until 2011. Because the period of 1995–2011 fully overlaps with the working period of QSCAT (1999–2009), comparing the merged radar signal against the ERS-2 data in 1995–2011 is the most direct validation available. We found concordant monthly dynamics between the merged radar signals and the ERS-2 signals during 1995–2011, with Pearson r value ranging from 0.79 to 0.98 across regions. These results evidenced that our merged radar data have a consistent C-band signal dynamic. The CScat data set (https://doi.org/10.6084/m9.figshare.20407857, Tao et al. 2022a) is expected to advance our understanding of the long-term changes in, e.g., global vegetation and soil moisture. The data set will be updated on a regular basis.&emsp

The 2015/16 El Niño brought severe drought and record-breaking temperatures in the tropics. Here, using satellite-based L-band microwave vegetation optical depth, we mapped changes of above-ground biomass (AGB) during the drought and in subsequent years up to 2019. Over more than 60% of drought-affected intact forests, AGB reduced during the drought, except in the wettest part of the central Amazon, where it declined 1 y later. By the end of 2019, only 40% of AGB reduced intact forests had fully recovered to the predrought level. Using random-forest models, we found that the magnitude of AGB losses during the drought was mainly associated with regionally distinct patterns of soil water deficits and soil clay content. For the AGB recovery, we found strong influences of AGB losses during the drought and of γ . γ is a parameter related to canopy structure and is defined as the ratio of two relative height (RH) metrics of Geoscience Laser Altimeter System (GLAS) waveform data—RH25 (25% energy return height) and RH100 (100% energy return height; i.e., top canopy height). A high γ may reflect forests with a tall understory, thick and closed canopy, and/or without degradation. Such forests with a high γ ( γ ≥ 0.3) appear to have a stronger capacity to recover than low- γ ones. Our results highlight the importance of forest structure when predicting the consequences of future drought stress in the tropics.

International audience; Ant communities are extremely diverse and provide a wide variety of ecological functions in tropical forests. Here we investigated the abiotic factors driving ant composition turnover across an elevational gradient at Mont Itoupé, French Guiana. Mont Itoupé is an isolated mountain whose top is covered by cloud forests, a biogeographical rarity that is likely to be threatened according to climate change scenarios in the region. We examined the influence of six soil, climatic and LiDAR derived vegetation structure variables on leaf-litter ant assembly (267 species) across nine 0.12-ha plots disposed at three elevations (ca. 400, 600 and 800m asl). We tested (a) whether species cooccurring within a same plot or a same elevation were more similar in terms of taxonomic, functional and phylogenetic composition, than species from different plots/elevations, and (b) which environmental variables significantly explained compositional turnover among plots. We found that the distribution of species and traits of ant communities along the elevational gradient was significantly explained by a turnover of environmental conditions, particularly in soil phosphorus and sand content, canopy height and mean annual relative humidity of soil. Our results shed light on the role exerted by environmental filtering in shaping ant community assembly in tropical forests. Identifying the environmental determinants of ant species distribution along tropical elevational gradients could help predicting the future impacts of global warming on biodiversity organization in vulnerable environments such as cloud forests.

International audience; AbstractKey messageWe used lightweight terrestrial laser scanning (TLS) to detect over 3000 stems per hectare across a 12-ha permanent forest plot in French Guiana, 81% of them

International audience; Vegetation optical depth (VOD) retrieved from microwave remote sensing techniques has been used as an important proxy for monitoring the vegetation dynamics at large scales. In a first study we developed a new X-VOD indice from inversion of the X-MEB model from the Microwave Scanning Radiometer 2 (AMSR2, 2012-present) observations. The Advanced Microwave Scanning Radiometer for EOS (AMSR-E, 2002–2011) and AMSR2 share many characteristics, such as local crossing time, incidence angle, frequency at X-band, etc, which makes it possible to produce a consistent long-term X-VOD product over a long period spanning from 2002 to present. The objective of this study was to merge the AMSR-E and AMSR2 observations in order, in a second step, to produce a consistent long-term X-VOD product spanning over the combined periods of AMSR-E and AMSR2.The main challenges in retrieving a consistent X-VOD data set from AMSR-E and AMSR2 are that (i) there is a bias between observations from the two sensors; and (ii) the lack of overlapping observations between the two sensors makes it impossible to achieve a direct inter-calibration. Here, to overcome this problem, we used AMSR-E slow rotation data (AMSR-E L1S), which has similar characteristics as AMSR-E and provided observations concurrently with AMSR2, as a bridge to calibrate the AMSR-E brightness temperature (TB) observations with AMSR2 TB.As our main objective in this TB calibration study was to produce a consistent long-term AMSR-E/AMSR2 X-VOD product, we evaluated the retrieved X-VOD product based on different inter-calibration, either global-based or pixel-based, methods. Both AMSR-E and AMSR2 X-VOD were evaluated against the Aboveground Biomass (AGB), Leaf Area Index (LAI) and the Normalized Difference Vegetation Index (NDVI). The results suggest that global-based inter-calibration methods using homogeneous and temporally stable land covers (Evergreen Broadleaf Forests and Snow and Ice) as reference calibration data sets provided the best results. For instance, the spatial relationships between X-VOD and AGB/LAI/NDVI are highly consistent over the AMSR-E and AMSR2 periods after the calibration work. This study laid a solid foundation for monitoring the dynamics of X-VOD, as a proxy of AGB, over the combined periods of AMSR-E and AMSR2 sensors (almost 20 years).

International audience; ESA's Soil Moisture Ocean Salinity (SMOS, since 2009) and NASA's Soil Moisture Active Passive (SMAP, since 2015) are the only two space-borne L-band radiometer missions currently in orbit, which provide key information on global surface soil moisture (SM) and vegetation water content (via the vegetation optical depth, VOD). However, to date very few studies considered merging SMOS and SMAP for both SM and VOD retrievals simultaneously. This study presents the first global long-term and continuous SM and L-band VOD (L-VOD) dataset retrieved after merging the SMOS and SMAP brightness temperature (TB) observations, called the SMOS-SMAP-INRAE-BORDEAUX or SMOSMAP-IB product. We first developed a fitted SMOS TB dataset at a fixed incidence angle of 40°, and next applied a monthly linear rescaling of SMAP TB to SMOS TB for each polarization to produce a merged SMOS/SMAP TB (θ = 40°) dataset. The retrievals were then based on a mono-angular retrieval algorithm sharing a similar forward model with the SMOS-IC and the official SMOS retrieval algorithms. Results showed that the inter-calibration approach we used here could effectively remove the bias between the SMAP TB and fitted SMOS TB, with bias values reduced to 0.01 K (−0.02 K) compared to 3.45 K (1.65 K) for V (H) polarization before inter-calibration. The SMOSMAP-IB SM and L-VOD retrievals based on this new inter-calibrated SMOS/SMAP TB led to metrics that were equally good or better than those of other products (i.e., ESA CCI, SMOS-IC and the official SMAP products). When considering only long duration products, SMOSMAP-IB SM retrievals exhibited (i) the highest overall median R value of 0.72 with in-situ data from ISMN (International Soil Moisture Network) during 2013–2018, followed by SMOS-IC (R = 0.68) and CCI (R = 0.67), and (ii) the same smallest ubRMSD values as CCI (ubRMSD = 0.057 m3/m3 vs 0.061 m3/m3 for SMOS-IC). L-VOD retrievals from SMOSMAP-IB were found to have comparable spatial and temporal skills to SMOS-IC. Spatially, they both correlated well with aboveground biomass (R = 0.87), and temporally, they both showed a good representation of the short vegetation NDVI signal and of the forest area loss in the Brazilian Amazon from 2011 to 2019. Developing SMOSMAP-IB is a step forward towards building a time-continuous L-band SM and VOD products in response to the possible failure of one of the SMOS or SMAP sensors in the future.

Lakes have played a critical role in providing water and ecosystem services for people and other organisms in China for millennia. However, accelerating climate change and economic boom have resulted in unprecedented changes in these valuable lakes. Using Landsat images covering the entity of the country, we explored the changes in China’s lakes and the associated driving forces over the last 30 years (i.e. mid-1980s to 2015). We discovered that China’s lakes have changed with divergent regional trends: in the sparsely populated Tibetan Plateau, lakes are abundant and the lake area has increased dramatically from 38 596 to 46 831 km2 (i.e. increased by 8235 km2, or 21.3%), whereas, in the densely populated northern and eastern regions, lakes are relatively scarce and the lake area has decreased from 36 659 to 33 657 km2 (i.e. decreased by 3002 km2, or 8.2%). In particular, severe lake decreases occurred in the Mongolia-Xinjiang Plateau and the Eastern Plain (−2151 km2). Statistical analyses indicated that climate was the most important factor controlling lake changes in the Tibetan Plateau, the Yun-Gui Plateau and the Northeast Plain. However, the strength of climatic control on lake changes was low in the Eastern Plain and the Mongolia-Xinjiang Plateau, where human activities, e.g. impoldering, irrigation and mining, have caused serious impacts on lakes. Further lake changes will exacerbate regional imbalances between lake resources and population distribution, and thus may increase the risk of water-resource crises in China.

Aims Plant height is a key functional trait related to aboveground biomass, leaf photosynthesis and plant fitness. However, large-scale geographical patterns in community-average plant height (CAPH) of woody species and drivers of these patterns across different life forms remain hotly debated. Moreover, whether CAPH could be used as a predictor of ecosystem primary productivity is unknown. Methods We compiled mature height and distributions of 11 422 woody species in eastern Eurasia, and estimated geographic patterns in CAPH for different taxonomic groups and life forms. Then we evaluated the effects of environmental (including current climate and historical climate change since the Last Glacial Maximum (LGM)) and evolutionary factors on CAPH. Lastly, we compared the predictive power of CAPH on primary productivity with that of LiDAR-derived canopy-height data from a global survey. Important Findings Geographic patterns of CAPH and their drivers differed among taxonomic groups and life forms. The strongest predictor for CAPH of all woody species combined, angiosperms, all dicots and deciduous dicots was actual evapotranspiration, while temperature was the strongest predictor for CAPH of monocots and tree, shrub and evergreen dicots, and water availability for gymnosperms. Historical climate change since the LGM had only weak effects on CAPH. No phylogenetic signal was detected in family-wise average height, which was also unrelated to the tested environmental factors. Finally, we found a strong correlation between CAPH and ecosystem primary productivity. Primary productivity showed a weaker relationship with CAPH of the tallest species within a grid cell and no relationship with LiDAR-derived canopy height reported in the global survey. Our findings suggest that current climate rather than historical climate change and evolutionary history determine the geographical patterns in CAPH. However, the relative effects of climatic factors representing environmental energy and water availability on spatial variations of CAPH vary among plant life forms. Moreover, our results also suggest that CAPH can be used as a good predictor of ecosystem primary productivity.

No waters left untouched We are increasingly aware of human impacts on biodiversity across our planet, especially in terrestrial and marine systems. We know less about fresh waters, including large rivers. Su et al. looked across such systems globally, focusing on several key measures of fish biodiversity. They found that half of all river systems have been heavily affected by human activities, with only very large tropical river basins receiving the lowest levels of change. Fragmentation and non-native species have also led to the homogenization of rivers, with many now containing similar species and fewer specialized lineages. Science , this issue p. 835

Lakes provide valuable water resources for people, life, and socioeconomic development in Northwestern China---a core region of the arid Central Asia where water is seriously deficient. In this study, we report the first investigation of the 30-year changes in all lakes larger than 1 km2 in Northwestern China and the associated driving forces. We found that the number of lakes increased from 121 to 135, and the total lake area increased from 5495 km2 to 6445 km2 in this region. However, increases in lake area occurred mainly in the less populated mountainous areas caused by glacier melting and increase in precipitation. Conversely, lakes in the densely populated regions decreased by 40% since the year of 2000, which was significantly related to the increasing intensity of irrigation. These results not only suggest an ongoing water crisis in Northwest China but also reflect the lake dynamics in arid Central Asia where climate changes and irrigation are occurring prevalently with high intensity. Given the fact that the benefits of glacier melt in Central Asia can be substantially offset by intensive irrigation, we thus call an urgent need for a policy of sustainable water management in this critical region.

Modeling the radiative transfer (RT) in heterogeneous forest scenes is important for understanding biophysical processes, as well as retrieving information from remotely sensed data. LiDAR (Light Detection and Ranging) is capable of providing highly detailed three-dimensional (3D) canopy structural information that can be used to parameterize RT models. In previous studies, point cloud data (such as terrestrial LiDAR data) are often voxelized with coarse resolutions, and the foliage voxels are often assumed to be turbid medium. In this study we propose a new voxel-based RT model, namely VBRT, that uses high resolution solid voxels to approximate 3D structure of forest more accurately than coarse resolution turbid medium voxels used in previous studies. Parallel computing techniques are used to speed up computation and the model can run on high performance computing (HPC) platforms. VBRT was tested in four virtual forest scenes, using the well-known physically based ray tracer (PBRT) as a benchmark. The Discrete Anisotropic Radiative Transfer (DART) model, which is based on turbid medium voxels, was also compared. Experimental results show that simulated digital imagery and bi-directional reflectance factor (BRF) by VBRT and PBRT are in good agreement, and the difference in simulation results can be reduced by using higher resolution voxels or larger number of samples per pixel. According to our test, parameterizing VBRT using high resolution terrestrial LiDAR data with 0.02 m voxels can produce more accurate results than DART with turbid medium voxels (0.1 m), although VBRT is more computation-intensive due to the use of higher resolution voxels. Our results indicate that VBRT has good potential in modeling radiation transfer in forests, as it is possible to parameterize the model using high density point cloud data such as terrestrial LiDAR data.

Terrestrial laser scanning (TLS) is a promising tool for estimating leaf area index (LAI). However, very few studies have considered the effect of clumping index Ω in the calculation of “true” LAI. In this study, we developed a new point cloud slicing method based on different incident zenith angles θ and retrieved the gap fraction using multiple-return information to obtain more accurate “true” LAI estimations. In addition, we described a new Ω retrieval method based on the gap size analysis theory to correct the effect of foliage occlusion. Ground validation data were collected by destructively sampling 35 trees and measuring all their leaves. Results show that the TLS-based “true” LAI estimations based on a single TLS scan are strongly correlated with the destructively sampled LAI measurements (R2 = 0.76, RMSE = 0.47). Moreover, our Ω retrieval method can effectively correct the effect of foliage occlusion. Other factors, such as the slicing resolution, percentage of laser beams with multiple returns, and scanning distance, have little effect on the final LAI estimation.

Dynamic global vegetation models are useful tools for the simulation of global carbon cycle. However, most models are hampered by the poor availability of global aboveground biomass (AGB) data, which is necessary for the model calibration process. Here, taking the integrated biosphere simulator model (IBIS) as an example, we evaluated the modeled carbon dynamics, including gross primary production (GPP) and potential AGB, at the global scale. The IBIS model was constrained by both in situ GPP and plot-level AGB data collected from the literature. Model results showed that IBIS could reproduce GPP with acceptable accuracy in monthly and annual scales. At the global scale, the IBIS-simulated total AGB was similar to those obtained in other studies. However, discrepancies were observed between the model-derived and observed AGB for pan-tropical forests. The bias in modeled AGB was mainly caused by the unchanged parameters over the global scale for a specific plant functional type. This study also showed that different meteorological inputs can introduce substantial differences in modeled AGB in the global scale, although this difference is small compared with parameter-induced differences. The conclusions of our research highlight the necessity of considering the heterogeneity of key model physiological parameters in modeling global AGB.

Woody residence time (τw) is an important parameter that expresses the balance between mature forest recruitment/growth and mortality. Using field data collected from the literature, this study explored the global forest τw and investigated its influence on model simulations of aboveground biomass (AGB) at a global scale. Specifically, τw was found to be related to forest age, annual temperature, and precipitation at a global scale, but its determinants were different among various plant function types. The estimated global forest τw based on the filed data showed large spatial heterogeneity, which plays an important role in model simulation of AGB by a dynamic global vegetation model (DGVM). The τw could change the resulting AGB in tenfold based on a site-level test using the Monte Carlo method. At the global level, different parameterization schemes of the Integrated Biosphere Simulator using the estimated τw resulted in a twofold change in the AGB simulation for 2100. Our results highlight the influences of various biotic and abiotic variables on forest τw. The estimation of τw in our study may help improve the model simulations and reduce the parameter's uncertainty over the projection of future AGB in the current DGVM or Earth System Models. A clearer understanding of the responses of τw to climate change and the corresponding sophisticated description of forest growth/mortality in model structure is also needed for the improvement of carbon stock prediction in future studies.

Terrestrial laser scanning (TLS) is a promising tool, used to retrieve leaf area index (LAI). However, the accuracy of LAI estimations based on TLS is still difficult to validate, because...

Spatial scale and pattern play important roles in forest aboveground biomass (AGB) estimation in remote sensing. Changes in the accuracy of satellite images-estimated forest AGBs against spatial scales and pixel distribution patterns has not been evaluated, because it requires ground-truth AGBs of fine resolution over a large extent, and such data are difficult to obtain using traditional ground surveying methods. We intend to quantify the accuracy of AGB estimation from satellite images on changing spatial scales and varying pixel distribution patterns, in a typical mixed coniferous forest in Sierra Nevada mountains, California. A forest AGB map of a 143 km2 area was created using small-footprint light detection and ranging. Landsat Thematic Mapper images were chosen as typical examples of satellite images, and resampled to successively coarser resolutions. At each spatial scale, pixels forming random, uniform, and clustered spatial patterns were then sampled. The accuracies of the AGB estimation based on Landsat images associated with varying spatial scales and patterns were finally quantified. The changes in the accuracy of AGB estimation from Landsat images are not monotonic, but increase up to 60–90 m in spatial scale, and then decrease. Random and uniform spatial patterns of pixel distributions yield better accuracy for AGB estimation than clustered spatial patterns. The corrected NDVI (NDVIc) was the best predictor of AGB estimation. A spatial scale of 60–90 m is recommended for forest AGB estimation at the Sierra Nevada mountains using Landsat images and those with similar spectral resolutions.

The global forest ecosystem, which acts as a large carbon sink, plays an important role in modeling the global carbon balance. An accurate estimation of the total forest carbon stock in the aboveground biomass (AGB) is therefore necessary for improving our understanding of carbon dynamics, especially against the background of global climate change. The forest area of China is among the top five globally. However, because of limitations in forest AGB mapping methods and the availability of ground inventory data, there is still a lack in the nationwide wall-to-wall forest AGB estimation map for China. In this study, we collected over 8000 ground inventory records from published literatures, and developed an AGB mapping method using a combination of these ground inventory data, Geoscience Laser Altimeter System (GLAS)/Ice, Cloud, and Land Elevation Satellite (ICESat) data, optical imagery, climate surfaces, and topographic data. An uncertainty field model was introduced into the forest AGB mapping procedure to minimize the influence of plot location uncertainty. Our nationwide wall-to-wall forest AGB mapping results show that the forest AGB density in China is 120 Mg/ha on average, with a standard deviation of 61 Mg/ha. Evaluation with an independent ground inventory dataset showed that our proposed method can accurately map wall-to-wall forest AGB across a large landscape. The adjusted coefficient of determination (R2) and root-mean-square error between our predicted results and the validation dataset were 0.75 and 42.39 Mg/ha, respectively. This new method and the resulting nationwide wall-to-wall forest AGB map will help to improve the accuracy of carbon dynamic predictions in China.

The rapid development of light detection and ranging (LiDAR) techniques is advancing ecological and forest research. During the last decade, numerous single tree segmentation techniques have been developed using airborne LiDAR data. However, accurate crown segmentation using terrestrial or mobile LiDAR data, which is an essential prerequisite for extracting branch level forest characteristics, is still challenging mainly because of the difficulties posed by tree crown intersection and irregular crown shape. In the current work, we developed a comparative shortest-path algorithm (CSP) for segmenting tree crowns scanned using terrestrial (T)-LiDAR and mobile LiDAR. The algorithm consists of two steps, namely trunk detection and subsequent crown segmentation, with the latter inspired by the well-proved metabolic ecology theory and the ecological fact that vascular plants tend to minimize the transferring distance to the root. We tested the algorithm on mobile-LiDAR-scanned roadside trees and T-LiDAR-scanned broadleaved and coniferous forests in China. Point-level quantitative assessments of the segmentation results showed that for mobile-LiDAR-scanned roadside trees, all the points were classified to their corresponding trees correctly, and for T-LiDAR-scanned broadleaved and coniferous forests, kappa coefficients ranging from 0.83 to 0.93 were obtained. We believe that our algorithm will make a contribution to solving the problem of crown segmentation in T-LiDAR scanned-forests, and might be of interest to researchers in LiDAR data processing and to forest ecologists. In addition, our research highlights the advantages of using ecological theories as guidelines for processing LiDAR data.

Forests play a key role in the global carbon cycle, and forest above ground biomass (AGB) is an important indictor to the carbon storage capacity and the potential carbon pool size of a forest ecosystem. Accurate estimation of forest AGB has become increasingly important for a wide range of end-users. Although satellite remote sensing provides abundant observations to monitor forest coverage, validation of coarse-resolution AGB derived from satellite observations is difficult because of the scale mismatch between the footprints of satellite observations and field measurements. In this study, we use airborne Lidar to bridge the scale gaps between satellite-based and field-based studies, and evaluate satellite-derived indices to estimate regional forest AGB. We found that: (1) Lidar data can be used to accurately estimate forest AGB using tree height and tree quadratic height, (2) linear regression, among four tested models, achieve the best performance (R2 = 0.74; RMSE = 183.57 Mg/ha); (3) for MODIS-derived vegetation indices at varied spatial resolution (250–1000 m), accumulated NDVI, accumulated LAI, and accumulated FPAR could explain 53–74% variances of forest AGB, whereas accumulated NDVI derived from 1 km MODIS products gives higher R2 (74%) and lower RMSE (13.4 Mg/ha) than others. We conclude that Lidar data can be used to bridge the scale gap between satellite and field studies. Our results indicate that combining MODIS and Lidar data has the potential to estimate regional forest AGB.

International audience; Tropical forests are a key component of the global carbon cycle. Yet, there are still high uncertainties in forest carbon stock and flux estimates, notably because of their spatial and temporal variability across the tropics. Several upcoming spaceborne missions have been designed to address this gap. High-quality ground data are essential for accurate calibration/validation so that spaceborne biomass missions can reach their full potential in reducing uncertainties regarding forest carbon stocks and fluxes. The BIOMASS mission, a P-band SAR satellite from the European Space Agency (ESA), aims at improving carbon stock mapping and reducing uncertainty in the carbon fluxes from deforestation, forest degradation, and regrowth. In situ activities in support of the BIOMASS mission were carried out in French Guiana and Gabon during the TropiSAR and AfriSAR campaigns. During these campaigns, airborne P-band SAR, forest inventory, and lidar data were collected over six study sites. This paper describes the methods used for forest inventory and lidar data collection and analysis, and presents resulting plot estimates and aboveground biomass maps. These reference datasets along with intermediate products (e.g., canopy height models) can be accessed through ESA's Forest Observation System and the Dryad data repository and will be useful for BIOMASS but also to other spaceborne biomass missions such as GEDI, NISAR, and Tandem-L for calibration/validation purposes. During data quality control and analysis, prospects for reducing uncertainties have been identified, and this paper finishes with a series of recommendations for future tropical forest field campaigns to better serve the remote sensing community.