No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Estimating aboveground biomass of urban trees based on ICESat-2 LiDAR and Zhuhai-1 hyperspectral data
... Random forest is a machine learning algorithm that exhibits high accuracy and robustness, making it suitable for analyzing highdimensional and highly correlated data sets (Wei et al., 2024). The algorithm operates on the principle of utilizing the bootstrap method to iteratively and randomly sample from the original data, constructing decision trees for each sample. ...
Estimation of forest biomass at regional scale based on GEDI spaceborne LiDAR data is of great significance for forest quality assessment and carbon cycle. To solve the problem of discontinuous data of GEDI footprints, this study mapped different echo indexes in the footprints to the surface by inverse distance weighted interpolation method, and verified the influence of different number of footprints on the interpolation results. Random forest algorithm was chosen to estimate the spruce-fir biomass combined with the parameters provided by GEDI and 138 spruce-fir sample plots in Shangri-La. The results show that: (1) By extracting different numbers of GEDI footprints and visualize it, the study revealed that a higher number of footprints correlates with a denser distribution and a more pronounced stripe phenomenon. (2) The prediction accuracy improves as the number of GEDI footprints decreases. The group with the highest R², lowest RMSE and lowest MAE was the footprint extracted every 100 shots, and the footprint extracted every 10 shots had the worst prediction effect. (3) The biomass of spruce-fir inverted by random forest ranged from 51.33 t/hm² to 179.83 t/hm², with an average of 101.98 t/hm². The total value was 3035.29 × 10⁴ t/hm². This study shows that the number and distribution of GEDI footprints will have a certain impact on the interpolation mapping to the surface information and presents a methodological reference for selecting the appropriate number of GEDI footprints to derive various vertical structure parameters of forest ecosystems.
... Remote sensing technology, as an efficient and non-destructive monitoring technology that can be easily applied over large areas, is widely used for precision agriculture crop monitoring. These include monitoring of biological variables [8], monitoring of plant pests and diseases [9], and yield prediction [10]. ...
Nitrogen is the main nutrient element in the growth process of white radish, and accurate monitoring of radish leaf nitrogen content (LNC) is an important guide for precise fertilization decisions for radish in the field. Using white radish LNC monitoring as an object, research on radish nitrogen hyperspectral estimation methods was carried out based on leaf hyperspectral and field sample nitrogen data at multiple growth stages using feature selection and integrated learning algorithm models. First, the Vegetation Index (VI) was constructed from hyperspectral data. We extracted sensitive features of hyperspectral data and VI response to radish LNC based on Pearson’s feature-selection approach. Second, a stacking-integrated learning approach is proposed using machine learning algorithms such as Support Vector Machine (SVM), Random Forest (RF), and Ridge and K-Nearest Neighbor (KNN) as the base model in the first layer of the architecture, and the Lasso algorithm as the meta-model in the second layer of the architecture, to realize the hyperspectral estimation of radish LNC. The analysis results show the following: (1) The sensitive bands of the radish LNC are mainly centered around 600–700 nm and 1950 nm, and the constructed sensitive VIs are also concentrated in this band range. (2) The Stacking model with spectral features as inputs achieved good prediction accuracy at the radish spectral leaf, with R2 = 0.7, MAE = 0.16, MSE = 0.05 estimated over the whole growth stage of radish. (3) The Lasso algorithm with variable filtering function was chosen as the meta-model, which has a redundant model-selection effect on the base model and helps to improve the quality of the integrated learning framework. This study demonstrates the potential of the stacking-integrated learning method based on hyperspectral data for spectral estimation of nitrogen content in radish at multiple growth stages.
This study used multi-platform remote sensing datasets to assess the impacts of land use and land cover (LULC) change on forest biomass density across the Three Gorges Reservoir (TGR) region along the Yangtze River. This study maps land use and land cover (LULC) change and forest aboveground biomass estimation (AGB) using Sentinel-2A and GEDI-L4B data based on observations between 2019 and 2022. The data analysis results of monitoring revealed a strong association frequency between land-use change and forest land cover since the significant relationship with forest aboveground biomass density has also been explored. The total deforestation decreased significantly in 2022. The accuracy assessment results show a classification accuracy of between 90.4% and 94.1%, with a kappa coefficient of 82.4%–92.1%. For the TGR region forest, the minimum to maximum mean above-ground biomass values are from 15.81 to 373.66 Mgha-1, respectively. Moreover, the fitting performance of the statistical analysis of aboveground biomass density of forest mean for the forest was quite decent with training (R²: 0.92, RMSE: 12.30 Mg/ha) and testing (R²: 0.79, RMSE: 18.07 Mg/ha) datasets. The results highlight LULC change impacts, particularly forest land class over forest AGB stands, and the effectiveness of the GEDI-L4B product in regional biomass estimation in the TGR region.
Urban land use/land cover (LULC) classification has long been a hotspot for remote sensing applications. With high spatio-temporal resolution and multispectral, the recently launched GF-6 satellite provides ideal open imagery for LULC mapping. In this study, we utilized multitemporal GF-6 images to generate six types of land features, including spectral bands, texture features, built-up, waterbody, vegetation, and red-edge indices. The minimum Redundancy Maximum Relevance (mRMR) algorithm was employed to optimize feature selection. Subsequently, Random Forest (RF) and Extreme Gradient Boosting (XGBT) were assessed using different feature selections. Besides, various feature configurations were designed for LULC classification and comparison. The results indicate that the mRMR-based RF method achieved the highest overall accuracy of 91.37%. The temporal red-edge indices were important features for urban LULC classification and contributed mainly to grassland and cropland. These results supplement existing classification methods and assist in improving LULC mapping in urban areas with complex landscapes.
Flash floods are among the world most destructive natural disasters, and developing optimum hybrid Machine Learning (ML) models for flash flood susceptibility (FFS) modeling remains a challenge. This study proposed novel intelligence algorithms based on a hybrid of several ensemble ML models (i.e., Bagged Flexible Discrim-inant Analysis (BAFDA), Extreme Gradient Boosting (XBG), Rotation Forest (ROF) and Boosted Generalized Additive Model (BGAM)) and wrapper-based factor optimization algorithms (i.e., Recursive Feature Elimination (RFE) and Boruta) to improve the accuracy of FFS mapping at Neka-Haraz watershed in Iran. In addition, the Random Search (RS) method is proposed for meta-optimization of the developed hybrid models hyper-parameters. This study considers 20 flash flood conditioning factors (CgFs) and 380 flood and non-flood locations to create a geospatial database. The performance of each hybrid model was evaluated by area under the receiver operating characteristic (ROC) curve (AUC) and several validation methods, such as efficiency. The developed hybrid models demonstrated good performance, with BGAM-Boruta achieving the highest performance (AUC = 0.953, and Efficiency = 0.910), followed by ROF-Boruta (AUC = 0.952), ROF-RFE (AUC = 0.951), BAFDA-Boruta (AUC = 0.950), BGAM-RFE (AUC = 0.950), ROF (AUC = 0.949), BGAM (AUC = 0.948), BAFDA-RFE (AUC = 0.943), XGB-Boruta (AUC = 0.943), BAFDA (AUC = 0.939), XGB-RFE (AUC = 0.938) and XGB (AUC = 0.911). In the BGAM-Boruta model, the regional coverage was about 46% for high to very high FFS areas. Moreover, the study revealed that distance to river, slope, rainfall, altitude, and distance to road CgFs are the most significant for FFS modeling in this region.
An accurate spatial distribution map of the urban dominant tree species is crucial for evaluating the ecosystem service value of urban forests and formulating urban sustainable development strategies. Spaceborne hyperspectral remote sensing has been utilized to distinguish tree species, but these hyperspectral data have a low spatial resolution (pixel size ≥ 30 m), which limits their ability to differentiate tree species in urban areas characterized by fragmented patches and robust spatial heterogeneity. Zhuhai-1 is a new hyperspectral satellite sensor with a higher spatial resolution of 10 m. This study aimed to evaluate the potential of Zhuhai-1 hyperspectral imagery for classifying the urban dominant tree species. We first extracted 32 reflectance bands and 18 vegetation indices from Zhuhai-1 hyperspectral data. We then used the random forest classifier to differentiate 28 dominant tree species in Shenzhen based on these hyperspectral features. Finally, we analyzed the effects of the classification paradigm, classifier, and species number on the classification accuracy. We found that combining the hyperspectral reflectance bands and vegetation indices could effectively distinguish the 28 dominant tree species in Shenzhen, obtaining an overall accuracy of 76.8%. Sensitivity analysis results indicated that the pixel-based classification paradigm was slightly superior to the object-based paradigm. The random forest classifier proved to be the optimal classifier for distinguishing tree species using Zhuhai-1 hyperspectral imagery. Moreover, reducing the species number could slowly improve the classification accuracy. These findings suggest that Zhuhai-1 hyperspectral data can identify the urban dominant tree species with accuracy and holds potential for application in other cities.
Forests play a crucial part in regulating global climate change since their aboveground biomass (AGB) relates to the carbon cycle, and its changes affect the main carbon pools. At present, the most suitable available SAR data for wall-to-wall forest AGB estimation are exploiting an L-band polarimetric SAR. However, the saturation issues were reported for AGB estimation using L-band backscatter coefficients. Saturation varies depending on forest structure. Polarimetric information has the capability to identify different aspects of forest structure and therefore shows great potential for reducing saturation issues and improving estimation accuracy. In this study, 121 polarimetric decomposition observations, 10 polarimetric backscatter coefficients and their derived observations, and six texture features were extracted and applied for forest AGB estimation in a tropical forest and a boreal forest. A parametric feature optimization inversion model (Multiple linear stepwise regression, MSLR) and a nonparametric feature optimization inversion model (fast iterative procedure integrated into a K-nearest neighbor nonparameter algorithm, KNNFIFS) were used for polarimetric features optimization and forest AGB inversion. The results demonstrated the great potential of L-band polarimetric features for forest AGB estimation. KNNFIFS performed better both in tropical (R2 = 0.80, RMSE = 22.55 Mg/ha, rRMSE = 14.59%, MA%E = 12.21%) and boreal (R2 = 0.74, RMSE = 19.82 Mg/ha, rRMSE = 20.86%, MA%E = 20.19%) forests. Non-model-based polarimetric features performed better compared to features extracted by backscatter coefficients, model-based decompositions, and texture. Polarimetric observations also revealed site-dependent performances.
Forest carbon is a large and uncertain component of the global carbon cycle. An important source of complexity is the spatial heterogeneity of vegetation vertical structure and extent, which results from variations in climate, soils, and disturbances and influences both contemporary carbon stocks and fluxes. Recent advances in remote sensing and ecosystem modeling have the potential to significantly improve the characterization of vegetation structure and its resulting influence on carbon. Here, we used novel remote sensing observations of tree canopy height collected by two NASA spaceborne lidar missions, Global Ecosystem Dynamics Investigation and ICE, Cloud, and Land Elevation Satellite 2, together with a newly developed global Ecosystem Demography model (v3.0) to characterize the spatial heterogeneity of global forest structure and quantify the corresponding implications for forest carbon stocks and fluxes. Multiple-scale evaluations suggested favorable results relative to other estimates including field inventory, remote sensing-based products, and national statistics. However, this approach utilized several orders of magnitude more data (3.77 billion lidar samples) on vegetation structure than used previously and enabled a qualitative increase in the spatial resolution of model estimates achievable (0.25° to 0.01°). At this resolution, process-based models are now able to capture detailed spatial patterns of forest structure previously unattainable, including patterns of natural and anthropogenic disturbance and recovery. Through the novel integration of new remote sensing data and ecosystem modeling, this study bridges the gap between existing empirically based remote sensing approaches and process-based modeling approaches. This study more generally demonstrates the promising value of spaceborne lidar observations for advancing carbon modeling at a global scale.
Synthetic aperture radar (SAR) backscatter based above-ground biomass (AGB) estimates are limited by the saturation of the backscatter-AGB curve. This work explores the potential of combining backscatter with polarimetric SAR interferometry (PolInSAR) estimated forest stand height for improved AGB estimation. The models combining L-band backscatter and TanDEM-X height are compared with established backscatter based models. The models are also temporally cross-validated, i.e., trained on one acquisition date and validated for other dates. It is observed that with the input of height, the combined models perform significantly better than backscatter based models, with an improvement in root mean square error (RMSE) between 19% and 46%. The model utilizing HV-polarized backscatter and TanDEM-X PolInSAR height provide the best case AGB inversion with an R2 = 0.78 and an RMSE of 27.1 Mg/ha or 22% of mean AGB. The results demonstrate the potential of the synergistic combination of L-band PolSAR (backscatter) and X-band PolInSAR (height) products for AGB mapping over a tropical forest range in India.
The objective of this study was to explore the effects of (1) the presence/absence of snow and snow depth, (2) solar noise, i.e., day/night and sun angle observations, and (3) strong/weak beam differences on ICESat-2 data in the context of data utility for forest AGB estimation. The framework of the study is multiphase modeling, where AGB field data and wall-to-wall airborne laser scanning (ALS) and Sentinel-2 data are used to produce proxy ALS plots on ICESat-2 track positions. Models between the predicted proxy AGB and the ICESat-2 photon data are then formulated and evaluated by subsets, such as only strong beam data captured in snowy conditions.
Our results indicate that, if possible, strong beam night data from snowless conditions should be used in AGB estimation, because our models showed clearly smallest RMSE (26.9%) for this data subset. If more data are needed, we recommend using only strong beam data and constructing separate models for the different data subsets. In the order of increasing RMSE%, the next best options were snow/night/strong (30.4%), snow/day/strong (33.5%), and snowless/day/strong (34.1%). Weak beam data from snowy night conditions could also be used if necessary (31.0%).
Accurate bathymetric and topographical information is crucial for coastal and marine applications. In the past decades, owing to its low cost and high efficiency, satellite-derived bathymetry has been widely used to estimate the depth of shallow water in coastal areas. However, insufficient spectral bands and availability of in situ water depths limit the application of satellite-derived bathymetry. Currently, the investigation about the bathymetric potential of hyperspectral imaging is relatively insufficient based on datasets of the Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2). In this study, Zhuhai-1 hyperspectral images and ICESat-2 datasets were utilized to perform nearshore bathymetry and explore the bathymetric capability by selecting different bands based on classical empirical models (the band ratio model and the linear band model). Furthermore, experimental results achieved at the South China Sea indicate that the combination of blue (2 and 3 band) and green (9 band) bands and the combination of red (10 and 12 band) and near-infrared (29 band) bands are most suitable to achieve nearshore bathymetry. Correspondingly, the highest accuracy of bathymetry reached root mean square error values of 0.98 m and 1.19 m for different band combinations evaluated through bathymetric results of reference water depth. The bathymetric accuracy of Zhuhai-1 image is similar with that of Sentinel-2 when employing the blue and green bands. The combination of red and near-infrared bands has a higher bathymetric accuracy for Zhuhai-1 image than that for Sentinel-2 image.
Accurate and effective mapping of forest aboveground biomass (AGB) in heterogeneous mountainous regions is a huge challenge but an urgent demand for resource managements and carbon storage monitoring. Conventional studies have related the plot-measured or LiDAR-based biomass to remote sensing data using pixel-based approaches. The object-based relationship between AGB and multi-source data from LiDAR, multi-frequency radar, and optical sensors were insufficiently studied. It deserves the further exploration that maps forest AGB using the object-based approach and combines LiDAR data with multi-sensor images, which has the smaller uncertainty of positional discrepancy and local heterogeneity, in heterogeneous mountainous regions. To address the improvement of mapping accuracy, satellite LiDAR data from GEDI and ICEsat-2, and images of ALOS-2 yearly mosaic L band SAR (Synthetic Aperture Radar), Sentinel-1 C band SAR, Sentinel-2 MSI, and ALOS-1 DSM were combined for pixel- and object-based forest AGB mapping in a vital heterogeneous mountainous forest. For the object-based approach, optimized objects during a multiresolution segmentation were acquired by the ESP (Estimation of the Scale Parameter) tool, and suitable predictors were selected using an algorithm named VSURF (Variable Selection Using Random Forests). The LiDAR variables at the footprint-level were extracted to connect field plots to the multi-sensor objects as a linear bridge. It was shown that forests’ AGB values varied by elevations with a mean value of 142.58 Mg/ha, ranging from 12.61 to 514.28 Mg/ha. The north slope with the lowest elevation (
Understanding the biomass, characteristics, and carbon sequestration of urban forests is crucial for maintaining and improving the quality of life and ensuring sustainable urban planning. Approaches to urban forest management have been incorporated into interdisciplinary, multifunctional, and technical efforts. In this review, we evaluate recent developments in urban forest research methods, compare the accuracy and efficiency of different methods, and identify emerging themes in urban forest assessment. This review focuses on urban forest biomass estimation and individual tree feature detection, showing that the rapid development of remote sensing technology and applications in recent years has greatly benefited the study of forest dynamics. Included in the review are light detection and ranging-based techniques for estimating urban forest biomass, deep learning algorithms that can extract tree crowns and identify tree species, methods for measuring large canopies using unmanned aerial vehicles to estimate forest structure, and approaches for capturing street tree information using street view images. Conventional methods based on field measurements are highly beneficial for accurately recording species-specific characteristics. There is an urgent need to combine multi-scale and spatiotemporal methods to improve urban forest detection at different scales.
Forest biomass plays an essential role in forest carbon reservoir studies, biodiversity protection, forest management, and climate change mitigation actions. Synthetic Aperture Radar (SAR), especially the polarimetric SAR with the capability of identifying different aspects of forest structure, shows great potential in the accurate estimation of total and component forest above-ground biomass (AGB), including stem, bark, branch, and leaf biomass. This study aims to fully explore the potential of polarimetric parameters at the C- and L-bands to achieve high estimation accuracy and improve the estimation of AGB saturation levels. In this study, the backscattering coefficients at different polarimetric channels and polarimetric parameters extracted from Freeman2, Yamaguchi3, H-A-Alpha, and Target Scattering Vector Model (TSVM) decomposition methods were optimized by a random forest algorithm, first, and then inputted into linear regression models to estimate the total forest AGB and biomass components of two test sites in China. The results showed that polarimetric observations had great potential in total and component AGB estimation in the two test sites; the best performances were for leaves at test site I, with R2 = 0.637 and RMSE = 1.27 t/hm2. The estimation of biomass components at both test sites showed obvious saturation phenomenon estimation according to their scatter plots. The results obtained at both test sites demonstrated the potential of polarimetric parameters in total and component biomass estimation.
Evapotranspiration, as a combination of evaporation and transpiration of water vapour, is a primary component of the global hydrological cycle. It accounts for significant losses of soil moisture from the earth to the atmosphere. Thus, reliable methods to monitor and forecast evapotranspiration are required for decision-making in many sectors. Reference evapotranspiration, denoted as ET, is a major parameter that is useful in quantifying soil moisture in a cropping system. This article aims to design a multi-stage deep learning hybrid Long Short-Term Memory (LSTM) predictive model that is coupled with Multivariate Empirical Mode Decomposition (MEMD) and Boruta-Random Forest (Boruta) algorithms to forecast ET in the drought-prone regions (i.e., Gatton, Fordsdale, Cairns) of Queensland, Australia. Daily data extracted from NASA’s Goddard Online Interactive Visualization and Analysis Infrastructure (GIOVANNI) and Scientific Information for Land Owners (SILO) repositories over 2003–2011 are used to build the proposed multi-stage deep learning hybrid model, i.e., MEMD-Boruta-LSTM, and the model’s performance is compared against competitive benchmark models such as hybrid MEMD-Boruta-DNN, MEMD-Boruta-DT, and a standalone LSTM, DNN and DT model. The test MEMD-Boruta-LSTM hybrid model attained the lowest Relative Root Mean Square Error (≤ 17%), Absolute Percentage Bias (≤ 12.5%) and the highest Kling-Gupta Efficiency (≥ 0.89) relative to benchmark models for all study sites. The proposed multi-stage deep hybrid MEMD-Boruta-LSTM model also outperformed all other benchmark models in terms of predictive efficacy, demonstrating its usefulness in the forecasting of the daily ET dataset. This MEMD-Boruta-LSTM hybrid model could therefore be employed in practical environments such as irrigation management systems to estimate evapotranspiration or to forecast ET.
Plain Language Summary
Estimation and mapping of forest biomass are vital to understand the role of forests in the global carbon budget and climate. Forest canopy height is an important indicator of biomass, carbon stock, and forest productivity and hence, incorporation of accurate forest canopy height information can result in improved estimation of forest biomass. The present study aims to map the forest canopy height and analyses the effect of including the canopy height information on aboveground biomass prediction by the synergistic use of multi‐sensor earth observation data using a machine learning algorithm. It presents a comprehensive methodology for forest canopy height and biomass estimation using active and passive remote sensing data. The developed approach can be implemented for mapping as well as monitoring the forest biomass/carbon stock and has huge relevance to carbon budgeting, climate conventions, and sustainable management of forests. The study was conducted using freely available satellite data, algorithm/software, and platform which makes it easily replicable in a time and cost‐effective manner.
With the advent of the next generation of space-based laser altimeters, ICESat-2 and GEDI, we are entering an exciting era of active remote sensing of forests that offers unprecedented opportunities for the observation of forest structure. Consistent comparisons of the accuracy of terrain and canopy height retrievals for these two missions are essential for continued improvement and further application. Because the time interval between the spaceborne products and validation data may introduce additional errors, we validate the newly released GEDI L2A product (version 2) and the ICESat-2 ATL08 product (version 4) using high-resolution, locally calibrated airborne lidar products acquired in the same year (2019) as the reference datasets. In addition, our study area contains 40 sites located in the U.S. mainland, Alaska, and Hawaii that encompass a variety of eco-climatic zones and vegetation cover types; thus, it avoids the uncertainties associated with small sample sizes and restricted spatial coverage. The results show that ICESat-2 and GEDI yield reasonable estimates of terrain height, with root mean squared errors (RMSEs) of 2.24 and 4.03 m for mid and low latitudes, respectively, and 0.98 m for high latitudes (ICESat-2 only). ICESat-2 outperforms GEDI across the board for terrain height retrieval, although they both have better accuracy than existing SRTM and GMTED DEM products. Analyses of the error factors suggest that steep slopes (>30 •) present the greatest challenge for both GEDI and ICESat-2; in addition, tall (>20 m) and dense canopies (>90%) forest ecosystems also reduce the accuracy of the terrain height estimates. When ICESat-2 and GEDI data are used for canopy height retrieval, the use of only strong/power beam data acquired at night is recommended, as the overall RMSEs decrease from 7.21 and 5.02 m to 3.93 and 3.56 m, respectively, compared to using all data regardless of daytime and beam strength. GEDI outperforms ICESat-2 across the board for canopy height retrieval, as ICESat-2 has a larger potential bias for almost all forest types and cover conditions. ICESat-2 tends to overestimate the canopy height of dwarf shrublands and underestimate the canopy height of forest, and there is a gradual downward shift in the distribution of residuals with increasing canopy height. Overall, ICESat-2 with photon counting technology and GEDI with full waveform technology each represent the state of the art in spaceborne laser altimeters for terrain and canopy height retrieval. Combined, these two missions can take advantage of the unique strengths of each instrument.
Forage supply of savanna grasslands plays a crucial role for local food security and consequently, a reliable monitoring system could help to better manage vital forage resources. To help installing such a monitoring system, we investigated whether in-situ hyperspectral data could be resampled to match the spectral resolution of multi- and hyperspectral satellites; if the type of sensor affected model transfer; and if spatio-temporal patterns of forage characteristics could be related to environmental drivers. We established models for forage quantity (green biomass) and five forage quality proxies (metabolisable energy, acid/neutral detergent fibre, ash, phosphorus). Hyperspectral resolution of the Hyperion satellite mostly resulted in higher accuracies (i.e. higher R², lower RMSE). When applied to satellite data, though, the greater quality of the multispectral Sentinel-2 satellite data leads to more realistic forage maps. By analysing a three-year time series, we found plant phenology and cumulated precipitation to be the most important environmental drivers of forage supply. We conclude that none of the investigated satellites provide optimal conditions for monitoring purposes. Future hyperspectral satellite missions like EnMAP, combining the high information level of Hyperion with the good data quality and resolution of Sentinel-2, will provide the prerequisites for installing a regular monitoring service.
Biomass is an important indicator in estimating the growth and yield of crops. Mastering the growth state of crops by evaluating and monitoring biomass is crucial. Correlation analysis was used to identify hyperspectral sensitive bands with the best correlation with biomass, and a hyperspectral vegetation index was constructed. Plant height data was coupled with hyperspectral reflectance and vegetation index data, and a biomass inversion model was constructed using partial least squares (PLS), support vector machine (SVM), and random forest (RF) regression analysis. The R², root mean square error (RMSE), and normalized root mean square error (NRMSE) were calculated to verify the accuracy of the model. Plant height coupled with hyperspectral reflectance and vegetation index data, R², RMSE, and NRMSE of the biomass model establishment and validation, compared with the hyperspectral data alone, improved model accuracy of the three algorithms for maize (Zea mays L.) biomass inversion coupled with plant height data with the relative improvement in accuracy of 13.58, 9.6, and 1.05%, respectively. Additionally, the model becomes more stable, overcoming the spectral saturation of canopy to a certain extent. It provides supporting information for accurate farmland planning, growth monitoring, and yield estimation.
Accurate registration is an essential prerequisite for analysis and applications involving remote sensing imagery. It is usually difficult to extract enough matching points for inter-band registration in hyperspectral imagery due to the different spectral responses for land features in different image bands. This is especially true for non-adjacent bands. The inconsistency in geometric distortion caused by topographic relief also makes it inappropriate to use a single affine transformation relationship for the geometric transformation of the entire image. Currently, accurate registration between spectral bands of Zhuhai-1 satellite hyperspectral imagery remains challenging. In this paper, a full-spectrum registration method was proposed to address this problem. The method combines the transfer strategy based on the affine transformation relationship between adjacent spectrums with the differential correction from dense Delaunay triangulation. Firstly, the scale-invariant feature transform (SIFT) extraction method was used to extract and match feature points of adjacent bands. The RANdom SAmple Consensus (RANSAC) algorithm and the least square method is then used to eliminate mismatching point pairs to obtain fine matching point pairs. Secondly, a dense Delaunay triangulation was constructed based on fine matching point pairs. The affine transformation relation for non-adjacent bands was established for each triangle using the affine transformation relation transfer strategy. Finally, the affine transformation relation was used to perform differential correction for each triangle. Three Zhuhai-1 satellite hyperspectral images covering different terrains were used as experiment data. The evaluation results showed that the adjacent band registration accuracy ranged from 0.2 to 0.6 pixels. The structural similarity measure and cosine similarity measure between non-adjacent bands were both greater than 0.80. Moreover, the full-spectrum registration accuracy was less than 1 pixel. These registration results can meet the needs of Zhuhai-1 hyperspectral imagery applications in various fields.
Forest canopy height is an indispensable forest vertical structure parameter for understanding the carbon cycle and forest ecosystem services. A variety of studies based on spaceborne Lidar, such as ICESat, ICESat-2 and airborne Lidar, were conducted to estimate forest canopy height at multiple scales. However, while a few studies have been conducted based on ICESat-2 simulated data from airborne Lidar data, few studies have analyzed ATL08 and ATL03 products derived from the ATLAS sensor onboard ICESat-2 for regional vegetation canopy height mapping. It is necessary and promising to explore how data obtained by ICESat-2 can be applied to estimate forest canopy height. This study proposes a new means to estimate forest canopy height, defined as the mean height of trees within a given forest area, using a combination of ICESat-2 ATL08 and ATL03 data and ZY-3 satellite stereo images. Five procedures were used to estimate the forest canopy height of the city of Nanning in China: (1) Processing ground photons in a 30 m × 30 m grid; (2) Extracting a digital surface model (DSM) using ZY-3 stereo images; (3) Calculating a discontinuous canopy height model (CHM) dataset; (4) Validating the DSM and ground photon height using GEDI data; (5) Estimating the regional wall-to-wall forest canopy height product based on the backpropagation artificial neural network (BP-ANN) model and Landsat 8 vegetation indices and independent accuracy assessments with field measured plots. The validation shows a root mean square error (RMSE) of 3.34 m to 3.47 m and a coefficient of determination R 2 = 0.51. The new method shows promise and can be used for large-scale forest canopy height mapping at various resolutions or in combination with other data, such as SAR images. Finally, this study analyzes resolutions and how to filter effective data when ATL08 data are directly used to generate regional or global vegetation height products, which will be the focus of future research.
National Aeronautics and Space Administration’s (NASA’s) Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) provides rich insights over the Earth’s surface through elevation data collected by its Advanced Topographic Laser Altimeter System (ATLAS) since its launch in September 2018. While this mission is primarily aimed at capturing ice measurements, ICESat-2 also provides data over vegetated areas, offering the capability to gain insights into ecosystem structure and the potential to contribute to the sustainable management of forests. This study involved an examination of the utility of ICESat-2 for estimating forest aboveground biomass (AGB). The objectives of this study were to: (1) investigate the use of canopy metrics for estimating AGB, using data extracted from an ICESat-2 transect over forests in south-east Texas; (2) compare the accuracy for estimating AGB using data from the strong beam and weak beam; and (3) upscale predicted AGB estimates using variables from Landsat multispectral imagery and land cover and canopy cover maps, to generate a 30 m spatial resolution AGB map. Methods previously developed with simulated ICESat-2 data over Sam Houston National Forest (SHNF) in southeast Texas were adapted using actual data from an adjacent ICESat-2 transect over similar vegetation conditions. Custom noise filtering and photon classification algorithms were applied to ICESat-2’s geolocated photon data (ATL03) for one beam pair, consisting of a strong and weak beam, and canopy height estimates were retrieved. Canopy height parameters were extracted from 100 m segments in the along-track direction for estimating AGB, using regression analysis. ICESat-2-derived AGB estimates were then extrapolated to develop a 30 m AGB map for the study area, using vegetation indices from Landsat 8 Operational Land Imager (OLI), National Land Cover Database (NLCD) landcover and canopy cover, with random forests (RF). The AGB estimation models used few canopy parameters and suggest the possibility for applying well-developed methods for modeling AGB with airborne light detection and ranging (lidar) data, using processed ICESat-2 data. The final regression model achieved a R2 and root mean square error (RMSE) value of 0.62 and 24.63 Mg/ha for estimating AGB and RF model evaluation with a separate test set yielded a R2 of 0.58 and RMSE of 23.89 Mg/ha. Findings provide an initial look at the ability of ICESat-2 to estimate AGB and serve as a basis for further upscaling efforts.
Human activities, such as research, innovation and industry, concentrate disproportionately in large cities. The ten most innovative cities in the United States account for 23% of the national population, but for 48% of its patents and 33% of its gross domestic product. But why has human activity become increasingly concentrated? Here we use data on scientific papers, patents, employment and gross domestic product, for 353 metropolitan areas in the United States, to show that the spatial concentration of productive activities increases with their complexity. Complex economic activities, such as biotechnology, neurobiology and semiconductors, concentrate disproportionately in a few large cities compared to less--complex activities, such as apparel or paper manufacturing. We use multiple proxies to measure the complexity of activities, finding that complexity explains from 40% to 80% of the variance in urban concentration of occupations, industries, scientific fields and technologies. Using historical patent data, we show that the spatial concentration of cutting-edge technologies has increased since 1850, suggesting a reinforcing cycle between the increase in the complexity of activities and urbanization. These findings suggest that the growth of spatial inequality may be connected to the increasing complexity of the economy. Balland et al. use data on scientific papers, patents, employment and GDP for 353 metropolitan areas in the United States to show that economic complexity drives the spatial concentration of productive activities in large cities.
In order to consider forest biomass as a real alternative for energy production,
it is critical to obtain accurate estimates of its availability using non-destructive
sampling methods. In this study, we estimate the biomass available in a
Scots pine-dominated forest (Pinus sylvestris L.) located in Spain. The biomass
estimates were obtained using LiDAR data combined with a multispectral camera
and allometric equations. The method used to fuse the data was based on
back projection, which assures a perfect match between both datasets. The
results present estimates for each of the seven different biomass components:
above ground, below ground, log, needles, and large, medium and small
branches. The accuracy of the models varied between R2 values of 0.46 and
0.67 with RMSE% ranging from 15.72% to 35.43% with all component estimates
below 20%, except for the model estimating biomass of big branches. The
models in this study are suitable for the estimation of biomass and demonstrate
that computation is possible at a fine scale for the different biomass
components. These remote sensing methods are sufficiently accurate to develop
biomass resource cartography for multiple energy uses.
The second batch of Zhuhai-1 microsatellites was successfully launched on 26 April 2018. The batch included four Orbita hyperspectral satellites (referred to as OHS-A, OHS-B, OHS-C, and OHS-D) and one video satellite (OVS-2A), which have excellent hyperspectral data acquisition abilities. For the first time in China, a number of hyperspectral satellite networks have been realized. To ensure the application of hyperspectral remote sensing data, a series of on-orbit geometry processing and accuracy verification studies has been carried out on the “Zhuhai-1” hyperspectral camera since the satellite was launched. This paper presents the geometric processing methods involved in the production of Zhuhai-1 hyperspectral satellite basic products, including geometric calibration and basic product production algorithms. The OHS images were used to perform on-orbit geometric calibration, and the calibration accuracy was better than 0.5 pixels. The registration accuracy of the image spectrum of the basic product after calibration, the single orientation accuracy, and the accuracy of the regional network adjustment were evaluated. The spectral registration accuracy of the OHS basic products is 0.3–0.5 pixels, which is equivalent to the spectral band calibration accuracy. The single orientation accuracy is better than 1.5 pixels and the regional network adjustment accuracy is better than 1.2 pixels. The generated area orthoimages meet the seamless edge requirements, which verifies that the OHS basic product image has good regional mapping capabilities and can meet the application requirements.
The main objective of this work was to study the feasibility of using the green red vegetation index (GRVI) and the red edge ratio (RE/R) obtained from UAS imagery for monitoring the effects of soil water deficit and for predicting fibre quality in a surface-irrigated cotton crop. The performance of these indices to track the effects of water stress on cotton was compared to that of the normalised difference vegetation index (NDVI) and crop water stress index (CWSI). The study was conducted during two consecutive seasons on a commercial farm where three irrigation frequencies and two nitrogen rates were being tested. High-resolution multispectral images of the site were acquired on four dates in 2017 and six dates in 2018, encompassing a range of matric potential values. Leaf stomatal conductance was also measured at the image acquisition times. At harvest, lint yield and fibre quality (micronaire) were determined for each treatment. Results showed that within each year, the N rates tested (> 180 kg N ha-1) did not have a statistically significant effect on the spectral indices. Larger intervals between irrigations in the less frequently irrigated treatments led to an increase (p < 0.05) in the CWSI and a reduction (p < 0.05) in the GRVI, RE/R, and to a lesser extent in the NDVI. A statistically significant and good correlation was observed between the GRVI and RE/R with soil matric potential and stomatal conductance at specific dates. The GRVI and RE/R were in accordance with the soil and plant water status when plants experienced a mild level of water stress. In most of the cases, the GRVI and RE/R displayed long-term effects of the water stress on plants, thus hampering their use for determinations of the actual soil and plant water status. The NDVI was a better predictor of lint yield than the GRVI and RE/R. However, both GRVI and RE/R correlated well (p < 0.01) with micronaire in both years of study and were better predictors of micronaire than the NDVI. This research presents the GRVI and RE/R as good predictors of fibre quality with potential to be used from satellite platforms. This would provide cotton producers the possibility of designing specific harvesting plans in the case that large fibre quality variability was expected to avoid discount prices. Further research is needed to evaluate the capability of these indices obtained from satellite platforms and to study whether these results obtained for cotton can be extrapolated to other crops.
The spectral properties of plant leaves relate to the state of their photosynthetic apparatus and the surrounding environment. An example is the well known photosynthetic downregulation, active on the time scale from minutes to hours, caused by reversible changes in the xanthophyll cycle pigments. These changes affect leaf spectral absorption and are frequently quantified using the photochemical reflectance index (PRI). This index can be used to remotely monitor the photosynthetic status of vegetation, and allows for a global satellite-based measurement of photosynthesis. Such earth observation satellites in near-polar orbits usually cover the same geographical location at the same local solar time at regular intervals. To facilitate the interpretation of these instantaneous remote PRI measurements and upscale them temporally, we measured the daily course of leaf PRI in two evergreen biomes—a European boreal forest and an Amazon rainforest. The daily course of PRI was different for the two locations: At the Amazonian forest, the PRI of Manilkara elata leaves was correlated with the average photosynthetic photon flux density (PPFD) (
,
) of the 40 minutes preceding the leaf measurement. In the boreal location, the variations in Pinus sylvestris needle PRI were only weakly (
,
) correlated with mean PPFD of the preceding two hours; for Betula pendula, the correlation was insignificant (
regardless of the averaging period. The measured daily PRI curves were specific to species and/or environmental conditions. Hence, for a proper interpretation of satellite-measured instantaneous photosynthesis, the scaling of PRI measurements should be supported with information on its correlation with PPFD.
While considerable research has focused on using either L-band or P-band SAR (Synthetic Aperture Radar) on their own for forest biomass retrieval, the use of the two bands simultaneously to improve forest biomass retrieval remains less explored. In this paper, we make use of L- and P-band airborne SAR and in situ data measured in the field together with laser scanning data acquired over one hemi-boreal (Remningstorp) and one boreal (Krycklan) forest study area in Sweden. We fit statistical models to different combinations of topographic-corrected SAR backscatter and forest heights estimated from PolInSAR for the biomass estimation, and evaluate retrieval performance in terms of R2 and using 10-fold cross-validation. The study shows that specific combinations of radar observables from L- and P-band lead to biomass predictions that are more accurate in comparison with single-band retrievals. The correlations and accuracies between the combinations of SAR features and aboveground biomass are consistent across the two study areas, whereas the retrieval performance varied for individual bands. P-band-based retrievals were more accurate than L-band for the hemi-boreal Remningstorp site and less accurate than L-band for the boreal Krycklan site. The aboveground biomass levels as well as the ground topography differ between the two sites. The results suggest that P-band is more sensitive to higher biomass and L-band to lower biomass forests. The forest height from PolInSAR improved the results at L-band in the higher biomass substantially, whereas no improvement was observed at P-band in both study areas. These results are relevant in the context of combining information over boreal forests from future low-frequency SAR missions such as the European Space Agency (ESA) BIOMASS mission, which will operate at P-band, and future L-band missions planned by several space agencies.
Cities throughout the world are advocating highly promoted tree plantings as a climate change mitigation measure. Assessing the carbon offsets associated with urban trees relative to other climate change policies is vital for sustainable development, planning, and solving environmental and socio-economic problems, but is difficult in developing countries. We estimated and assessed carbon dioxide (CO2) storage, sequestration, and emission offsets by public trees in the Medellin Metropolitan Area, Colombia, as a viable Nature-Based Solution for the Neotropics. While previous studies have discussed nature-based solutions and explored urban tree carbon dynamics in high income countries, few have been conducted in tropical cities in low-middle income countries, particularly within South America. We used a public tree inventory for the Metropolitan Area of the Aburrá Valley and an available urban forest functional model, i-Tree Streets, calibrated for Colombia's context. We found that CO2 offsets from public trees were not as effective as cable cars or landfills. However, if available planting spaces are considered, carbon offsets become more competitive with cable cars and other air quality and socio-economic co-benefits are also provided. The use of carbon estimation models and the development of relevant carbon accounting protocols in Neotropical cities are also discussed. Our nature-based solution approach can be used to better guide management of urban forests to mitigate climate change and carbon offset accounting in tropical cities lacking available information.
Forests sequester large quantity of carbon in their woody biomass and hence accurate estimation of forest biomass is extremely crucial. The present study aims at combining information from spaceborne LiDAR (ICESat/GLAS) and high resolution optical data to estimate forest biomass. Estimation of aboveground biomass (AGB) at ICESat/GLAS footprint level was done by integrating data from multiple sensors using two regression algorithms, viz. random forest (RF) and support vector machine (SVM). The study used forest height and canopy return ratio (rCanopy) for determination of effective size of ICESat/GLAS footprints for field data collection. The forest height was predicted with root mean square error (RMSE) of 1.35 m. The study showed that six most important parameters derived from LiDAR, and passive optical data were able to explain 78.7% (adjusted) variation in the observed AGB with an RMSE of 13.9 Mg ha–1. It was also observed that 15 most important parameters were able to explain 83% (adjusted) variation in the observed AGB. It was found that SVM regression algorithm explained 88.7% of variation in AGB with an RMSE of 13.6 Mg ha–1 on the combined datasets while RF regression algorithm explained 83.5% of variation in AGB with an RMSE of 20.57 Mg ha–1. The study demonstrated that RF regression algorithm performs equally well on datasets
Improving the urban environmental quality relies mainly on the increasing of urban forests capacity to store carbon dioxide. This study assesses the floristic diversity of urban institutional lands in Bukavu and their potential to reduce atmospheric CO2. An exhaustive inventory over three sites (Collège Alfajiri, Cathédrale Notre-Dame de la Paix and Institut Supérieur Pédagogique) of Bukavu led to the identification of 1 113 trees of which the diameter at breast height (1.30 m) ranged from 4.9 to 161 cm. Results reveal a floristic diversity made up of 4 families of conifers with 4 species and 14 of broadleaves with 21 species. Average densities were of 54 trees ha-1 and 5.21 m2 ha-1 of basal area. Urban-based allometric equations used yielded up to 312.8 tons of carbon stored in trees aboveground biomass equivalent to 1 147.9 tons of CO2 reduced from the atmosphere over the three sites. The rate of carbon storage reaches 15.1 tons ha-1. Thus, trees of the three institutional sites in Bukavu play an important role in reducing atmospheric CO2 and contribute, thereby, to mitigate global climate change effects. Given the current environmental challenge associated with high population growth rate in cities, the urban forest ecosystem in DRC requires to be extended and further
investigation.
Accurate estimates of forest aboveground biomass (AGB) could reduce uncertainties in the characterization of terrestrial carbon fluxes. Light Detection and Ranging (LiDAR) provides an accurate measure of canopy height and vertical structure and information for the estimation of aboveground biomass of vegetation. Spaceborne large footprint LiDAR (ICESat GLAS) acquires over 250 million observations over forest regions globally and has been used successfully for forest height and biomass estimation in various sites. In this paper, airborne LiDAR and ICESat GLAS data were used to estimate aboveground biomass of forest at footprint level in Yunnan, China. Vegetation products from EOS MODIS and ENVISAT MERIS were used to expand these discrete estimations from GLAS data to a wall-to-wall map. The R2 between ICESat GLAS waveform parameters and airborne LiDAR estimated forest AGB is 0.52 after training with 260 footprints. Results showed that the total forest AGB in Yunnan Province was 1272 million ton and the average was 94 Mg/ha. The amount and distribution of predicted aboveground biomass were in good agreement with the reference data. The results showed that the predict model using GLAS data could be used to estimate regional forest aboveground biomass successfully.
This article describes a R package Boruta, implementing a novel feature selection algorithm for finding all relevant variables. The algorithm is designed as a wrapper around a Random Forest classification algorithm. It iteratively removes the features which are proved by a statistical test to be less relevant than random probes. The Boruta package provides a convenient interface to the algorithm. The short description of the algorithm and examples of its application are presented.
The anthocyanin (Anth) content in leaves provides valuable information about the physiological status of plants. Thus, there is a need for accurate, efficient, practical methodologies to estimate this biochemical parameter. Reflectance measurement is a means of quickly and nondestructively assessing leaf Anth content in situ. The objective of this study was to test the overall performance and accuracy of nondestructive techniques for estimating Anth content in grapevine leaves. Relationships were established between Anth content and four vegetation indices: NIR (near-infrared)/green, red/green, anthocyanin reflectance index (ARI, based on reflectances in bands within the green and the red-edge regions), and a modified anthocyanin reflectance index (MARI, based on reflectances in green, red edge, and NIR). The algorithms for Anth retrieval were calibrated. The accuracy of Anth prediction was evaluated using an independent data set containing sampled leaves from two field-grown grape cultivars (Saint Croix and Saint Pepin) with no adjustment of the coefficients after initial calibration. Although Anth in the validation data set was widely variable, from 3 to 45 nmol cm-2, the ARI and MARI algorithms were capable of accurately predicting Anth content in grapevine leaves with a root mean square error below 3 nmol cm-2 and 2.3 nmol cm-2, respectively. Such an approach has potential for developing simple hand-held field instrumentation for accurate nondestructive Anth estimation and for analyzing digital airborne or satellite imagery to assist in making informed decisions regarding vineyard management.
This study evaluates and compares the performance of four machine learning classifiers-support vector machine (SVM), normal Bayes (NB), classification and regression tree (CART) and K nearest neighbor (KNN)-to classify very high resolution images, using an object-based classification procedure. In particular, we investigated how tuning parameters affect the classification accuracy with different training sample sizes. We found that: (1) SVM and NB were superior to CART and KNN, and both could achieve high classification accuracy (> 90%); (2) the setting of tuning parameters greatly affected classification accuracy, particularly for the most commonly-used SVM classifier; the optimal values of tuning parameters might vary slightly with the size of training samples; (3) the size of training sample also greatly affected the classification accuracy, when the size of training sample was less than 125. Increasing the size of training samples generally led to the increase of classification accuracies for all four classifiers. In addition, NB and KNN were more sensitive to the sample sizes. This research provides insights into the selection of classifiers and the size of training samples. It also highlights the importance of the appropriate setting of tuning parameters for different machine learning classifiers and provides useful information for optimizing these parameters.
Forest aboveground biomass (AGB) was mapped throughout China using large footprint LiDAR waveform data from the Geoscience Laser Altimeter System (GLAS) onboard NASA's Ice, Cloud, and land Elevation Satellite (ICESat), Moderate Resolution Imaging Spectro-radiometer (MODIS) imagery and forest inventory data. The entire land of China was divided into seven zones according to the geographic characteristics of the forests. The forest AGB prediction models were separately developed for different forest types in each of the seven forest zones at GLAS footprint level from GLAS waveform parameters and biomass derived from height and diameter at breast height (DBH) field observation. Some waveform parameters used in the prediction models were able to reduce the effects of slope on biomass estimation. The models of GLAS-based biomass estimates were developed by using GLAS footprints with slopes less than 20° and slopes = 20°, respectively. Then, all GLAS footprint biomass and MODIS data were used to establish Random Forest regression models for extrapolating footprint AGB to a nationwide scale. The total amount of estimated AGB in Chinese forests around 2006 was about 12,622 Mt vs. 12,617 Mt derived from the seventh national forest resource inventory data. Nearly half of all provinces showed a relative error (%) of less than 20%, and 80% of total provinces had relative errors less than 50%.
In spite of considerable efforts to monitor global vegetation, biomass quantification in drylands is still a major challenge due to low spectral resolution and considerable background effects. Hence, this study examines the potential of the space-borne hyperspectral Hyperion sensor compared to the multispectral Landsat OLI sensor in predicting dwarf shrub biomass in an arid region characterized by challenging conditions for satellite-based analysis: The Eastern Pamirs of Tajikistan. We calculated vegetation indices for all available wavelengths of both sensors, correlated these indices with field-mapped biomass while considering the multiple comparison problem, and assessed the predictive performance of single-variable linear models constructed with data from each of the sensors. Results showed an increased performance of the hyperspectral sensor and the particular suitability of indices capturing the short-wave infrared spectral region in dwarf shrub biomass prediction. Performance was considerably poorer in the area with less vegetation cover. Furthermore, spatial transferability of vegetation indices was not feasible in this region, underlining the importance of repeated model building. This study indicates that upcoming space-borne hyperspectral sensors increase the performance of biomass prediction in the world’s arid environments.
The present study deals with the estimation of aboveground biomass (AGB) using synthetic aperture radar (SAR) data in a highly complex mangrove assemblage of Lothian Island, Indian Sundarbans. A comparative analysis between the SAR datasets was also performed to check whether the L band in ALOS-PALSAR-2 data has any advantages over the C band in RISAT-1 and Sentinel-1 data. Empirical equations available in the literature were used to calculate plot-level biomass, which was then regressed with backscatter values produced from SAR datasets to obtain an AGB map of Lothian Island. AGB in Lothian Island ranged from 1.15 t/ha to 79.83 t/ha with an average value of 20.23 t/ha. Backscatter generated from RISAT-1 RH, Sentinel VH and ALOS-PALSAR-2 HV and VH performed best when regressed with field observations with R2 values of 0.91, 0.83, 0.78 and 0.97 respectively. This research could be useful as a valuable input in carbon mapping and could serve as a baseline report of blue carbon stock availability on Lothian Island, Sundarbans.
Analysis ready data (ARD) has recently become a popular standard for medium-resolution satellite images. The most representative datasets, namely Sentinel-2 Level-2A products and Landsat Collection 2 Level-2 Science Products, have met ARD standards, with public access. The Chinese Gaofen (GF) series Level-1A wide field of view (WFV) product, which has a similar ground sampling distance (GSD) to Sentinel-2 Level-2A products, is open to the public for its scenes and rational function model (RFM). However, the lower geometric processing accuracy of WFV products of approximately 3-pixel geolocation deviation restricts their wide application and further development towards ARD. The causes of low accuracy are twofold, namely the impact of calibration errors and the inaccuracy of the exterior orientation parameters. In terms of improving the accuracy of RFM, bundle adjustment and the bias compensation model (BCM) are currently the most widely accepted methods. However, existing approaches do not account for the imaging characteristics of WFV scenes, which are only for narrow field-of-view scenes and do not help with nonlinear internal distortions. To address the shortcomings of these methods, a joint orientation method for WFV cameras is proposed in this study. Owing to the high level of geometric accuracy of Sentinel-2 Level-1C reference scenes, this study analyzed the expression of WFV camera attitude errors on the image side. A polynomial exterior orientation model suitable for WFV cameras is also provided. We performed a refined calibration of the nonlinear internal distortions using multi-temporal GF-6 Level-1A WFV images, because of the additional error introduced by the cross-track inconsistency of the Sentinel-2 reference data. The method developed in this study has been experimentally verified to achieve satisfactory geometric positioning accuracy with GF-6 WFV products from 3-pixel to 0.3-pixel (5 m). This study allows optimization of the geometric performance of GF-6 Level-1A WFV products to the level of Sentinel-2 Level-1C products, which is of considerable importance in developing GF WFV images to ARD.
Shenzhen is one of the world's cities with the fastest urbanization and the most astonishing changes, and it is also a typical benchmark of Chinese rapid urbanization miracle. Under the tide of neoliberalism and globalization, the central government gave Shenzhen the priority of reform and experimentation first, then it has been rapidly pushed to the front stage as a test field for China's reform and opening up. Shenzhen has become the world's first-tier city in a short time with its location advantage and high-quality talent attraction, as well as active private economy and scientific and technological innovation. Through the multi-dimensional data analysis of the great changes during Shenzhen's urbanization, and a summary of the advantages of the Shenzhen model, this paper systematically answers the question that why Shenzhen can achieve urbanization so quickly. Furthermore, it concludes challenges of Shenzhen's future development and discusses the inspirations of the Shenzhen model for other cities. Based on the above analysis, this paper constructs an organic framework comprehensively covering achievements, characteristics, challenges, and inspirations of Shenzhen's rapid urbanization model. In the future, Shenzhen needs to transform from “Shenzhen Speed” to “Shenzhen Quality” to realize sustainable high-quality urbanization.
Aboveground biomass (AGB) is an essential variable in the study of the carbon cycle, and remote sensing provides the only viable means for mapping and monitoring biomass at large scales in time and space. Still, inconsistencies in biomass estimates are large which necessitates methods to achieve consistent mapping of biomass over time and space. In this study, we combine Lidar-derived biomass-estimates from GLAS measurements, time series analysis of 20 years of Landsat data, and machine learning algorithms to map biomass for a study area in the Amazon basin. The results show that we are able to map AGB consistently over space and time at annual intervals between 1999 and 2019, with a root-mean-square error (RMSE) of 64 to 92 Mg AGB / ha (depending on the evaluation methodology used) for the best performing models, with AGB estimates ranging between 0 to 600 Mg AGB / ha . Models run with the Extreme gradient boosting (XGBoost) resulted in the lowest errors. The presented methodological approach makes it possible to map biomass over time in large areas with wide biomass ranges, and without saturation below 400 Mg AGB / ha.
The upcoming BIOMASS mission will provide P-band repeat-pass PolInSAR data from space for the improved mapping of global biomass. PolInSAR technique has been widely validated with the potential to invert forest height and estimate forest aboveground biomass (AGB). However, the robustness of PolInSAR-based AGB estimation across different sites still lacks full evaluation, especially for those with a varied forest type, heterogeneity (varied growth ratio between cover and height), and topographic relief. In this study, we concentrated on backscatter decomposition and forest height inversion, and developed a robust AGB estimation method that can be applied to different sites. Two dense and closed tropical forest sites (Paracou and Nouragues) and one open and heterogeneous boreal forest site (Krycklan) were selected as the study areas, and the corresponding airborne PolInSAR, LiDAR, and ground measured AGB data were used for validation and analysis. Results show that ground backscatter has the strongest correlation with AGB in boreal forests, but this correlation cannot be transferred to the tropical forests. Only canopy volume backscatter is almost free from topographic influence, and its relationship with AGB across three sites can be formulated using one exponential equation, producing the best estimation accuracy, with R² of 0.79 and RMSE of 61.5 tons/ha (relative RMSE of 20.0 %). Multi-baseline PolInSAR retrieved forest height with little bias in spite of the presence of temporal decorrelation. One power equation can be used to correlate PolInSAR forest height with AGB across three different sites, and LOO (leave-one-out) validation shows the R² of 0.85 and RMSE of 51.8 tons/ha (relative RMSE of 16.9 %). However, the RVoG-inverted PolInSAR FH was found to mainly represent the top forest height for open and heterogeneous forests, which means PolInSAR FH (forest height) lacks consideration for forest horizontal structure (e.g. forest density). In contrast, volume backscatter better captured forest density, and the proposed AGB model that combines PolInSAR FH and volume backscatter further improved the AGB estimation accuracy, especially for open forests: the plot-scale validation from all three sites shows R² was improved from 0.79 (volume backscatter) and 0.85 (PolInSAR FH) to 0.89, and RMSE decreased from 61.5 and 51.8 to 45.2 (relative RMSE of 14.7 %) tons/ha; for region-scale validation, R² was improved from 0.77 and 0.83 to 0.89, and RMSE decreased from 64.2 (relative RMSE of 39.0 %) and 54.5 (34.5 %) to 48.1 (29.4 %) tons/ha.
Accurately mapping carbon stocks of urban trees is necessary for urban managers to design strategies to mitigate climate change. However, the aboveground carbon stocks of urban trees are usually underestimated by passive remote sensing data because of the signal saturation problem. The research is the first attempt to develop a framework to map aboveground carbon density of trees in urban areas by synergizing Ice, Cloud and Land Elevation Satellite-2 (ICESat-2) LiDAR data with Gaofen-2 (GF-2) imagery. The framework consists of three key steps. First, we used a support vector machine classifier to classify GF-2 images and extracted urban tree regions. Second, we estimated the tree carbon density of ICESat-2 strips by developing a ICESat-2 photon feature-based aboveground carbon density estimation model. Third, we mapped the carbon density of urban trees by developing a synergistic model between ICESat-2 and GF-2 data based on an object-oriented method. We tested the approach for the areas within the fifth ring road of Beijing, China. The results showed that the 50th percentile height (PH50) of nighttime photons was a good predictor for estimating carbon density of urban trees, with a R² of 0.69 and a Root Mean Square Error (RMSE) of 2.81 kg C m⁻². Using the spectral features generated by GF-2 imagery, we could further extrapolate the carbon density estimated by ICESat-2 strip data to a full coverage of accurate mapping carbon density by urban trees, resulting in a R² of 0.64 and a RMSE of 2.32 kg C m⁻². The carbon stocks within the fifth ring road of Beijing were 8.28×108 kg in total, with the mean carbon density of 3.52 kg C m⁻². Such estimations were larger than that of previous study using passive remote sensing data only, suggesting the integration of spaceborne LiDAR and spectral data could greatly reduce the underestimation of carbon stocks of urban trees. Our approach can more accurately estimate carbon stocks of urban trees and has the potential to be applicable in other cities.
For structural health monitoring, electrical resistivity measurement (ERM) method is commonly employed for the detection of concrete's durability, as indicated by the chloride permeability and the corrosion of steel reinforcement. However, according to previous experimental studies, ERM results are susceptible to significant uncertainties due to multiple influencing factors such as concrete water/cement ratio and structure curing environment as well as their complex interrelationships. The present study therefore proposes an XGBoost algorithm-based prediction model which considers all potential influential factors simultaneously. A database containing 800 experimental instances composed of 16 input attributes is constructed according to existing reported studies and utilized for training and testing the XGBoost model. Statistical scores (RMSE, MAE and R²) and the GridsearchCV feature are applied to evaluate and optimize the established model respectively. Results show that the proposed XGBoost model achieves satisfactory predictive performance as demonstrated by high coefficients of regression fitting lines (0.991 and 0.943) and comparatively low RMSE values (4.6 and 11.3 kΩ·cm) for both training and testing sets respectively. The analyses of the attribute importance ranking also reveal that curing age and cement content have the greatest influence on ERM results.
The Ice, Cloud, and land Elevation Satellite - 2 (ICESat-2) observatory was launched on 15 September 2018 to measure ice sheet and glacier elevation change, sea ice freeboard, and enable the determination of the heights of Earth's forests. ICESat-2's laser altimeter, the Advanced Topographic Laser Altimeter System (ATLAS) uses green (532 nm) laser light and single-photon sensitive detection to measure time of flight and subsequently surface height along each of its six beams. In this paper, we describe the major components of ATLAS, including the transmitter, the receiver and the components of the timing system. We present the major components of the ICESat-2 observatory, including the Global Positioning System, star trackers and inertial measurement unit. The ICESat-2 Level 1B data product (ATL02) provides the precise photon round-trip time of flight, among other data. The ICESat-2 Level 2A data product (ATL03) combines the photon times of flight with the observatory position and attitude to determine the geodetic location (i.e. the latitude, longitude and height) of the ground bounce point of photons detected by ATLAS. The ATL03 data product is used by higher-level (Level 3A) surface-specific data products to determine glacier and ice sheet height, sea ice freeboard, vegetation canopy height, ocean surface topography, and inland water body height.
A local-scale mesquite tree (Prosopis glandulosa Torr.) aboveground biomass map contribute to our understanding of the spatial distribution of woody plant aboveground biomass, and carbon stocks and fluxes in rangeland ecosystems. The objective of the study was examining a methodological approach to use airborne lidar data and multispectral imagery to create very high spatial resolution local-scale mesquite tree aboveground biomass maps by comparing three statistical methods and identifying significant prediction variables. The three statistical methods were the stepwise regression, the least absolute shrinkage and selection operator (LASSO), and the random forests. These methods were applied to establish the mesquite tree aboveground biomass equations and model from the in-situ mesquite tree aboveground biomass with the lidar metrics and multispectral data. The results showed the stepwise regression and LASSO had limited adj-R² and MSE. However, the random forests method with combined multispectral imagery and lidar data presented acceptable MSE and R² (1.08 Mg ha⁻¹ and 0.37). In summary, the random forests method with combined multispectral imagery and lidar data offered the most reliable and reasonable combination to generate a very high spatial resolution local-scale mesquite tree aboveground biomass map.
Existing ecological service assessments of urban green space have concentrated on natural environment, while research on the social services function remains underexplored. Taking Shenzhen as an example, this study starts with integrating the function of green space system with human needs. An indicator assessment technique that combines spatialization, rasterization and detailed of land ecological assessment was established with regard to the benefits of landscape aesthetics, function of disaster prevention and mitigation, and accessibility of park greenbelts. The results show that Shenzhen has very high recreational and cultural value, as the areas above medium level account for 66% of the city’s land area, and the recreational and cultural services are worth approximately 40.8million Yuan. Forest parks and comprehensive parks are important places to carry out social service function, particularly the disaster prevention and mitigation function, and the green spaces for disaster prevention and mitigation can be found in most of the disaster-prone areas. The accessibility of service shows a certain level of centrality, since the highest grade is concentrated in the central Bao’an, central Longgang, and the central part of the city. Based on the GIS overlay analysis, this study recognizes the distribution of essential patches, and constructs the structure of essential patches in urban green greenbelt systems by a combination of point, line, and surface elements. It shows the services of the Shenzhen greenbelt system have high potential for social integration, and this study also discusses the policy implication for macro decision-making. The assessment results objectively reflect multiple social services of the green space system in this region and provide a reference for the management, planning and construction of urban ecology in similar cities.
The Landsat 8 mission provides new opportunities for quantifying the distribution of above-ground carbon at moderate spatial resolution across the globe, and in particular drylands. Furthermore, coupled with structural information from space-based and airborne laser altimetry, Landsat 8 provides powerful capabilities for large-area, long-term studies that quantify temporal and spatial changes in above-ground biomass and cover. With the planned launch of ICESat-2 in 2017 and thus the potential to couple Landsat 8 and ICESat-2 data, we have unprecedented opportunities to address key challenges in drylands, including quantifying fuel loads, habitat quality, biodiversity, carbon cycling, and desertification.
Synthetic aperture radar (SAR) is one of the most promising remote sensors to map forest carbon. The unique spaceborne and long-wavelength SAR data currently available are L-band data, but their relationship with forest biomass is still controversial, particularly for high biomass values. While many studies assume a complete loss of sensitivity above a saturation point, typically around 100 t.ha− 1, others assume a continuous positive correlation between SAR backscatter and biomass. The objective of this paper is to revisit the relationship between L-band SAR backscatter and dense tropical forest biomass for a large range of biomass values using both theoretical and experimental approaches. Both approaches revealed that after reaching a maximum value, SAR backscatter correlates negatively with forest biomass. This phenomenon is interpreted as a signal attenuation from the forest canopy as the canopy becomes denser with increasing biomass. This result has strong implications for L-band vegetation mapping because it can lead to a greater-than-expected under-estimation of biomass. The consequences for L-band biomass mapping are illustrated, and solutions are proposed.
