Multisource Remote Sensing Data Research Articles

Drought Monitoring of Spring Maize in the Songnen Plain Using Multi-Source Remote Sensing Data

Agricultural production is highly susceptible to the impact of drought. How to improve agricultural drought-monitoring capability has always been a research hotspot. Based on multi-source remote-sensing data, a novel comprehensive drought index (CDI) for spring maize was developed using the random forest model, and its feasibility was tested by using agricultural drought indices and agricultural statistics in this study. Then, the spatiotemporal characteristics of spring maize drought in the Songnen Plain from 2001 to 2018 were evaluated using the CDI. The results showed that: (1) the CDI effectively monitored spring maize drought in the Songnen Plain, outperforming other drought indices. (2) The monitoring results indicated that spring maize in the Songnen Plain was affected by large-scale droughts in 2001, 2004, 2007, and 2017, which was consistent with national drought disaster statistics. (3) By changing the drought barycenter, the drought barycenter of spring maize generally tended to the south and west of the Songnen Plain, so drought-prevention measures should be strengthened in these areas in the future. While factors affecting crop yield extended beyond drought, the variations in spring maize yield indirectly reflected the effectiveness of drought monitoring in this study.

Formation and Hazard Analysis of Landslide Damming Based on Multi-Source Remote Sensing Data

Remote sensing plays an increasingly important role in the investigation of natural hazards, not only by obtaining specific data related to hazards, but also by realizing targeted research by combining with other data and/or technologies. Small-scale landslide hazard chain events occur frequently in mountainous areas with fragile geological environments and have strong destructive effects, yet have been somewhat understudied. This paper analyzes the Zhoujiaba (ZJB) landslide hazard chain that occurred in Longnan City on 18 August 2020. On the basis of the comprehensive application of multi-source remote sensing data, combined with time-series InSAR technology, electrical resistivity tomography (ERT), and numerical simulations, we studied the formation mechanism, damming characteristics, and potential outburst scenarios of this event. Our research suggests that geological structure and strong natural weathering are the preconditions for landslide development, which is eventually induced by extreme rainfall. Specific topographic conditions determine the rapid sliding and accumulation of landslide materials, and ultimately result in river damming. Our simulation results showed that a flood, rather than a debris flow, will be the result of dam outburst. When the simulated upstream inflow is 1.5 times that when the landslide occurred, 68% of the downstream village area will be flooded. The artificial spillway can effectively reduce the scale of the potential outburst flood, but there remains a risk of dam failure owing to the shallow depth. Our study of the hazard chain of a small-scale landslide using a combination of methods will provide a valuable reference for the analysis and treatment of similar hazard chains.

Enhancing Flood Simulation in Data-Limited Glacial River Basins through Hybrid Modeling and Multi-Source Remote Sensing Data

Due to the scarcity of observational data and the intricate precipitation–runoff relationship, individually applying physically based hydrological models and machine learning (ML) techniques presents challenges in accurately predicting floods within data-scarce glacial river basins. To address this challenge, this study introduces an innovative hybrid model that synergistically harnesses the strengths of multi-source remote sensing data, a physically based hydrological model (i.e., Spatial Processes in Hydrology (SPHY)), and ML techniques. This novel approach employs MODIS snow cover data and remote sensing-derived glacier mass balance data to calibrate the SPHY model. The SPHY model primarily generates baseflow, rain runoff, snowmelt runoff, and glacier melt runoff. These outputs are then utilized as extra inputs for the ML models, which consist of Random Forest (RF), Gradient Boosting (GDBT), Long Short-Term Memory (LSTM), Deep Neural Network (DNN), Support Vector Machine (SVM) and Transformer (TF). These ML models reconstruct the intricate relationship between inputs and streamflow. The performance of these six hybrid models and SPHY model is comprehensively explored in the Manas River basin in Central Asia. The findings underscore that the SPHY-RF model performs better in simulating and predicting daily streamflow and flood events than the SPHY model and the other five hybrid models. Compared to the SPHY model, SPHY-RF significantly reduces RMSE (55.6%) and PBIAS (62.5%) for streamflow, as well as reduces RMSE (65.8%) and PBIAS (73.51%) for floods. By utilizing bootstrap sampling, the 95% uncertainty interval for SPHY-RF is established, effectively covering 87.65% of flood events. Significantly, the SPHY-RF model substantially improves the simulation of streamflow and flood events that the SPHY model struggles to capture, indicating its potential to enhance the accuracy of flood prediction within data-scarce glacial river basins. This study offers a framework for robust flood simulation and forecasting within glacial river basins, offering opportunities to explore extreme hydrological events in a warming climate.

Quantitative Characterization of Coastal Cliff Retreat and Landslide Processes at Portonovo–Trave Cliffs (Conero, Ancona, Italy) Using Multi-Source Remote Sensing Data

The integration of multiple data sources, including satellite imagery, aerial photography, and ground-based measurements, represents an important development in the study of landslide processes. The combination of different data sources can be very important in improving our understanding of geological phenomena, especially in cases of inaccessible areas. In this context, the study of coastal areas represents a real challenge for the research community, both for the inaccessibility of coastal slopes and for the numerous drivers that can control coastal processes (subaerial, marine, or endogenic). In this work, we present a case study of the Conero Regional Park (Northern Adriatic Sea, Ancona, Italy) cliff-top retreat, characterized by Neogenic soft rocks (flysch, molasse). In particular, the study is focused in the area between the beach of Portonovo and Trave (south of Ancona), which has been studied using aerial orthophoto acquired between 1978 and 2021, Unmanned Aerial Vehicle (UAV) photographs (and extracted photogrammetric model) surveyed in September 2021 and 2012 LiDAR data. Aerial orthophotos were analyzed through the United States Geological Survey’s (USGS) tool Digital Shoreline Analysis System (DSAS) to identify and estimate the top-cliff erosion. The results were supported by the analysis of wave data and rainfall from the correspondent period. It has been found that for the northernmost sector (Trave), in the examined period of 40 years, an erosion up to 40 m occurred. Furthermore, a Digital Elevation Model (DEM) of Difference (DoD) between a 2012 Digital Terrain Model (DTM) and a UAV Digital Surface Model (DSM) was implemented to corroborate the DSAS results, revealing a good agreement between the retreat areas, identified by DSAS, and the section of coast characterized by a high value of DoD.

Comparing Forest Understory Fuel Classification in Portugal Using Discrete Airborne Laser Scanning Data and Satellite Multi-Source Remote Sensing Data

Wildfires burn millions of hectares of forest worldwide every year, and this trend is expected to continue growing under current and future climate scenarios. As a result, accurate knowledge of fuel conditions and fuel type mapping are important for assessing fire hazards and predicting fire behavior. In this study, 499 plots in six different areas in Portugal were surveyed by ALS and multisource RS, and the data thus obtained were used to evaluate a nationwide fuel classification. Random Forest (RF) and CART models were used to evaluate fuel models based on ALS (5 and 10 pulse/m2), Sentinel Imagery (Multispectral Sentinel 2 (S2) and SAR (Synthetic Aperture RaDaR) data (C-band (Sentinel 1 (S1)) and Phased Array L-band data (PALSAR-2/ALOS-2 Satellite) metrics. The specific goals of the study were as follows: (1) to develop simple CART and RF models to classify the four main fuel types in Portugal in terms of horizontal and vertical structure based on field-acquired ALS data; (2) to analyze the effect of canopy cover on fuel type classification; (3) to investigate the use of different ALS pulse densities to classify the fuel types; (4) to map a more complex classification of fuel using a multi-sensor approach and the RF method. The results indicate that use of ALS metrics (only) was a powerful way of accurately classifying the main four fuel types, with OA = 0.68. In terms of canopy cover, the best results were estimated in sparse forest, with an OA = 0.84. The effect of ALS pulse density on fuel classification indicates that 10 points m−2 data yielded better results than 5 points m−2 data, with OA = 0.78 and 0.71, respectively. Finally, the multi-sensor approach with RF successfully classified 13 fuel models in Portugal, with moderate OA = 0.44. Fuel mapping studies could be improved by generating more homogenous fuel models (in terms of structure and composition), increasing the number of sample plots and also by increasing the representativeness of each fuel model.

Regional Monitoring of Leaf ChlorophyII Content of Summer Maize by Integrating Multi-Source Remote Sensing Data

This study addresses the problem of restricted ability for large-scale monitoring due to the limited cruising time of unmanned aerial vehicles (UAV) by identifying an optimal leaf ChlorophyII content (LCC) inversion machine learning model at different scales and under different parameterization schemes based on simultaneous observations of ground sampling, UAV flight, and satellite imagery. The following results emerged: (1) The correlation coefficient between most remote sensing features (RSFs) and LCC increased as the remote scale expanded; thus, the scale error caused by the random position difference between GPS and measuring equipment should be considered in field sampling observations. (2) The LCC simulation accuracy of the UAV multi-spectral camera using four machine learning algorithms was ExtraTree > GradientBoost > AdaBoost > RandomForest, and the 20- and 30-pixel scales had better accuracy than the 10-pixel scale, while the accuracy for three feature combination schemes ranked combination of extremely significantly correlated RSFs > combination of significantly correlated and above RSFs > combination of all features. ExtraTree was confirmed as the optimal model with the feature combination of scheme 2 at the 20-pixel scale. (3) Of the Sentinel-2 RSFs, 27 of 28 were extremely significantly correlated with LCC, while original band reflectance was negatively correlated, and VIs were positively correlated. (4) The LCC simulation accuracy of the four machine learning algorithms ranked as ExtraTree > GradientBoost > RandomForest > AdaBoost. In a comparison of two parameterization schemes, scheme 1 had better accuracy, while ExtraTree was the best algorithm, with 11 band reflectance as input RSFs; the RMSE values for the training and testing data sets of 0.7213 and 1.7198, respectively.

Dust Outbreaks across East Iran: Application of Multi-Source Remote Sensing Data (AMSR-E and FengYun3-MWRI) on the Effects of Soil Moisture

One of the most significant hydro-meteorological and agricultural variables is soil moisture, yet measuring it remains a difficult task. Due to the significant spatial fluctuation of soil moisture, it is difficult to quantify it in a particular spot or field across a sizable region. Despite the thermal band's limitations in assessing soil moisture, MODIS and AVHRR, which are inappropriate were utilized in this investigation. The study examined the impact of soil moisture on dust outbreak. Soil moisture in the study domain was monitored using field techniques and the hybrid model. It combined multi-sourced remote sensing data, obtained from AMSER-E and FY-3 satellites. AMSER-E satellite measures the light temperature in five frequencies ranging from 6. 9 to 89 GHz based on data obtained from AMSER-E. Findings revealed areas with a spatial scale of 25 km2 has a 12-hour time step or variability in dust storm, thereby influencing soil moisture content within the zone of study. In addition to introducing acceptable potentials of the passive microwave band for accurate and applied monitoring of the soil moisture, the present results are viewed as a reliable source for studies on drought in time scale. The study shows that Zabol in Sistan has the highest annual average of 80.7 dust storm days. Soil moisture estimates serve a great deal for preparing soil moisture maps and the evaluation of temporal and spatial variations of soil moisture in study region to address issues related to dust storms.
 In order to identify the areas affected by dust storms and understand how dust particles are dispersed in the Sistan region, satellite image processing was employed using MODIS 1 sensor images obtained from the TERRA satellite.

Vegetation Growth Monitoring in Jiaozhou Bay Coastal Wetland Based on Multi-Source Remote Sensing Data

Jiaozhou Bay coastal wetland, located in the mouth of the Dagu River, is an estuary wetland. In this wetland seawater blends with freshwater. It is a rare type of national wetland, a unique ecosystem on earth. It is of great research value, but there are few articles conducting multi-indicator studies about Jiaozhou Bay coastal wetland ecosystem. This paper helps to formulate wetland conservation measures comprehensively through multi-source remote sensing monitoring. Based on NDVI, FVC, LAI, NPP and the spatial distribution changes of land use, this paper is a synthetical exploration of the vegetation growth of Jiaozhou Bay coastal wetland. The results show that: 1) From 2013 to 2019, NDVI, LAI and NPP indices rose to varying degrees in the coastal wetlands of Jiaozhou Bay. FVC showed a broad downward trend, and its indicators are declining in the north-central region; 2) The major land types are impervious surface and grassland, where the decrease proportion of grassland is the largest, and the increase proportion of impervious surface is the largest in the study area; 3) Analytical results demonstrate that the overall vegetation situation of Jiaozhou Bay coastal wetland improved on the whole, but with local deterioration. So, ecological protection measures in the study area have begun to bear fruit.

River Ecological Protection and Restoration Using Multi-source Remote Sensing Data

River Ecological Protection and Restoration Using Multi-source Remote Sensing Data

Understanding Spatial-Temporal Interactions of Ecosystem Services and Their Drivers in a Multi-Scale Perspective of Miluo Using Multi-Source Remote Sensing Data

In the face of rapid urbanization and global climate change, understanding the trade-offs and synergies of wetland city ecosystem services is vital for mitigating regional ecological and environmental risks, and enhancing human well-being. The Dongting Lake Basin is an ecologically fragile area of global significance. Uncontrolled resource utilization and intensive human activities have severely damaged the ecological environment, including in Miluo. Thus, it is of paramount research importance to uncover the trade-offs and synergies of ecosystem services and their driving mechanisms in Miluo. To achieve this, we classified Miluo’s land use data over the past two decades using a random forest model and Landsat imagery. We quantified the major ecosystem services in Miluo by employing ecological process models such as InVEST, RUSLE, and CASA. Additionally, we examined the trade-offs and synergies between ecosystem services at different scales and identified the driving mechanisms using multi-source remote sensing data. The results revealed that forests exhibited the highest level of ecosystem services, while urban ecosystem services experienced a significant decline. Over the past two decades, Miluo displayed notable trade-offs and synergies between ecosystem services, with synergies prevailing as the dominant pattern, particularly at the county scale. Furthermore, human activities emerged as the primary driver of changes in Miluo’s ecosystem services during the 20-year period. Therefore, it is imperative for scientists, policymakers, and civil society to develop effective and scientifically sound strategies to mitigate the ecological risks resulting from rapid urbanization and climate change in the future.

Monitoring of Land Cover and Vegetation Changes in Juhugeng Coal Mining Area Based on Multi-Source Remote Sensing Data

Coal is the most prevalent energy source in China and plays an important role in ensuring energy security. The continuous monitoring of coal mining activities is helpful to clarify the incremental space of coal production and establish a rational framework for future coal production capacity. In this study, a multi-source remote sensing approach utilizing SPOT 4, GF, and Landsat data is employed to monitor land cover and vegetation changes in the Juhugeng mining area of the Muli coalfield over a span of nearly 20 years. The analysis incorporates an object-oriented classification method and a vegetation parameter to derive insights. The findings reveal that the mining operations can be divided into two periods, since their initiation in 2003 until their cessation in 2021, with a dividing point around 2013/2014. The initial phase witnessed rapid and even accelerated expansion of the mine, while the subsequent phase was characterized by more stable development and the implementation of some restorative measures for the mine environment. Although the vegetation parameter, Fractional Vegetation Cover (FVC), indicates some reclamation efforts within the mining area, the extent of the reclaimed land remains limited. This study demonstrates the effective application of object-oriented classification in conjunction with the vegetation parameter FVC for monitoring coal mining areas.

A spatiotemporal collaborative approach for precise crop planting structure mapping based on multi-source remote-sensing data

ABSTRACT In order to ensure food security, it is crucial to collect agricultural information efficiently and accurately. Remote sensing has become increasingly important in obtaining crop distribution information on a large scale. However, current research based on satellite platforms struggles to meet the requirements of high-precision and large-scale crop monitoring simultaneously. To address this challenge, we propose a method for achieving fine-scale crop classification by integrating remote-sensing data from various satellite platforms by constructing temporal-scale crop features within the parcels using Sentinel-2A, Landsat-8, and Gaofen-6. We adopt a feature-matching method to fill in missing values in the time-series feature construction process, to avoid issues with unidentifiable crops. The classification results of the Yellow River basin of the Ningxia region show that our method can achieve a wide range of crop discrimination on a fine scale, with an overall accuracy of 80%. Our proposed method demonstrates the potential of integrating multi-platform remote-sensing data to achieve fine-scale crop classification, which can aid decision-making for farmers, government agencies, and other stakeholders involved in the agricultural sector.

Spatial-Temporal Characteristics and Driving Forces of Aboveground Biomass in Desert Steppes of Inner Mongolia, China in the Past 20 Years

The desert steppe serves as a transitional zone between grasslands and deserts, and long-term monitoring of aboveground biomass (AGB) in the desert steppe is essential for understanding grassland changes. While AGB observation techniques based on multisource remote-sensing data and machine-learning algorithms have been widely applied, research on monitoring methods specifically for the desert steppe remains limited. In this study, we focused on the desert steppe of Inner Mongolia, China, as the study area and used field sampling data, MODIS data, MODIS-based vegetation indices (VI), and environmental factors (topography, climate, and soil) to compare the performance of four commonly used machine-learning algorithms: multiple linear regression (MLR), partial least-squares regression (PLS), random forest (RF), and support vector machine (SVM) in AGB estimation. Based on the optimal model, the spatial–temporal characteristics of AGB from 2000 to 2020 were calculated, and the driving forces of climate change and human activities on AGB changes were quantitatively analyzed using the random forest algorithm. The results are as follows: (1) RF demonstrated outstanding performance in terms of prediction accuracy and model robustness, making it suitable for AGB estimation in the desert steppe of Inner Mongolia; (2) VI contributed the most to the model, and no significant difference was found between soil-adjusted VIs and traditional VIs. Elevation, slope, precipitation, and temperature all had positive effects on the model; (3) from 2000 to 2020, the multiyear average AGB in the study area was 58.34 g/m2, exhibiting a gradually increasing distribution pattern from the inner region to the outer region (from north to south); (4) from 2000 to 2020, the proportions of grassland with AGB slightly and significantly increasing trend in the study area were 87.08% and 5.13%, respectively, while the proportions of grassland with AGB slightly and significantly decreasing trend were 7.76% and 0.05%, respectively; and (5) over the past 20 years, climate change, particularly precipitation, has been the primary driving force behind AGB changes of the study area. This research holds reference value for improving desert steppe monitoring capabilities and the rational planning of grassland resources.

Reconstruction of all-weather land surface temperature based on a combined physical and data-driven model.

At present, the remote sensing (RS) thermal infrared (TIR) images that are commonly used to obtain land surface temperature (LST) are contaminated by clouds and thus cannot obtain spatiotemporal integrity of LST. To solve this problem, this study combined a physical model with strong interpretability with a data-driven model with high data adaptability. First, the physical model (Weather Research and Forecast (WRF) model) was used to generate LST source data. Then, combined with multisource RS data, a data-driven method (random forest (RF)) was used to improve the accuracy of the LST, and a model framework for a data-driven auxiliary physical model was formed. Finally, all-weather MODIS-like data with a spatial resolution of 1km were generated. Beijing, China, was used as the study area. The results showed that in cases of more clouds and fewer clouds, the reconstructed all-weather LST had a high spatial continuity and could restore the spatial distribution details of the LST well. The mean absolute error (MAE), root mean square error (RMSE), and correlation coefficient (ρ) in the case of more (fewer) clouds were ranked as follows: MAE < 1K (< 2K), RMSE < 2K (< 2K), and ρ > 0.9. The errors obeyed an approximately normal distribution. The total MAE, RMSE, and ρ were 0.80K, 1.09K, and 0.94K, respectively. Generally, the LST reconstructed in this paper had a high accuracy, and the model could provide all-weather MODIS-like LST to compensate for the disadvantages of satellite TIR images (i.e., contamination by clouds and inability to obtain complete LST values).

Dynamic Water Quality Changes in the Main Stream of the Yangtze River from Multi-Source Remote Sensing Data

Total nitrogen (TN) and total phosphorus (TP) are important indicators of water quality. Although water quality can be obtained with high accuracy using traditional measurement methods, the cost is high and the area is limited. In the past a single-satellite remote sensing system was normally used to estimate water quality at a large scale, while bands were fewer with limited accuracy. In this paper, inversion models for TN and TP are obtained and validated in the main stream of the Yangtze River using multi-source remote sensing data. The joint inversion models for TN and TP have higher accuracy (R2=0.81 and 0.86, RMSE=0.51 and 0.10 mg L−1) than the single-satellite inversion models (R2=0.61−0.62 and 0.59−0.75, RMSE=0.41−0.61 and 0.07−0.12 mg L−1). Using these models, water quality changes in the Yangtze River are obtained from 2019 to 2021. It is found that TN and TP in the upstream and downstream are high. In spring and autumn, the water quality is poor. The water quality in the Yangtze River is mostly Class III with improvement. Furthermore, it is found that TN and TP are negatively correlated with the water level, temperature and flow in Jiujiang. The p value between water quality and the water level is higher than for other factors, with −0.76 and −0.64 for TN and TP, respectively.

Inversion of Soil Moisture on Farmland Areas Based on SSA-CNN Using Multi-Source Remote Sensing Data

Soil moisture is a crucial factor in the field of meteorology, hydrology, and agricultural sciences. In agricultural production, surface soil moisture (SSM) is crucial for crop yield estimation and drought monitoring. For SSM inversion, a synthetic aperture radar (SAR) offers a trustworthy data source. However, for agricultural fields, the use of SAR data alone to invert SSM is susceptible to the influence of vegetation cover. In this paper, based on Sentinel-1 microwave remote sensing data and Sentinel-2 optical remote sensing data, a convolution neural network optimized by sparrow search algorithm (SSA-CNN) was suggested to invert farmland SSM. The feature parameters were first extracted from pre-processed remote sensing data. Then, the correlation analysis between the extracted feature parameters and field measured SSM data was carried out, and the optimal combination of feature parameters for SSM inversion was selected as the input data of the subsequent models. To enhance the performance of the CNN, the hyper-parameters of CNN were optimized using SSA, and the SSA-CNN model was built for SSM inversion based on the obtained optimal hyper-parameter combination. Three typical machine learning approaches, including generalized regression neural network, random forest, and CNN, were used for comparison to show the efficacy of the suggested method. With an average coefficient of determination of 0.80, an average root mean square error of 2.17 vol.%, and an average mean absolute error of 1.68 vol.%, the findings demonstrated that the SSA-CNN model with the optimal feature combination had a better accuracy among the 4 models. In the end, the SSM of the study region was inverted throughout four phenological periods using the SSA-CNN model. The inversion results indicated that the suggested method performed well in local situations.

Mapping and Influencing the Mechanism of CO2 Emissions from Building Operations Integrated Multi-Source Remote Sensing Data

Urbanization has led to rapid growth in energy consumption and CO2 emissions in the building sector. Building operation emissions (BCEs) are a major part of emissions in the building life cycle. Existing studies have attempted to estimate fine-scale BCEs using remote sensing data. However, there is still a lack of research on estimating long-term BCEs by integrating multi-source remote sensing data and applications in different regions. We selected the Beijing–Tianjin–Hebei (BTH) urban agglomeration and the National Capital Region of Japan (NCRJ) as research areas for this study. We also built multiple linear regression (MLR) models between prefecture-level BCEs and multi-source remote sensing data. The prefecture-level BCEs were downscaled to grid scale at a 1 km2 resolution. The estimation results verify the method’s difference and accuracy at different development stages. The multi-scale BCEs showed a continuous growth trend in the BTH urban agglomeration and a significant downward trend in the NCRJ. The decrease in energy intensity and population density were the main factors contributing to the negative growth of BCEs, whereas GDP per capita and urban expansion significantly promoted it. Through our methods and analyses, we contribute to the study of estimating greenhouse gas emissions with remote sensing and exploring the environmental impact of urban growth.

Long-Term Changes in Water Body Area Dynamic and Driving Factors in the Middle-Lower Yangtze Plain Based on Multi-Source Remote Sensing Data

The accurate monitoring of long-term spatial and temporal changes in open-surface water bodies offers important guidance for water resource security and management. In the middle and lower reaches of the Yangtze River, the monitoring of water body changes is especially critical due to the dense population and drastic climate change. Due to the complexity of the physical environment in which the water bodies are located, the advantages and disadvantages of various water body detection rules can vary in large-scale areas. In this paper, we use Landsat 5/7/8 data to extract the area of water bodies in the study area and analyze their spatial and temporal trends from 1984 to 2020 using the Google Earth Engine (GEE) platform. We propose an improved water body extraction rule based on an existing multi-indicator water body algorithm that combines impervious surface data and digital elevation model data. In this study, the performance of the improved algorithm was cross-validated using seven other water body indicator algorithms, and the results showed the following: (1) the rule accurately retained information about the water body while minimizing the interference of shadows on the extracted water body. (2) On the annual scale from 1984 to 2020, the open-surface water body dataset extracted using this improved rule showed that the turning point for the area of each water body type was 2011, with an overall decreasing trend in area before 2011 and an increasing trend in area after 2011, with the exception of special years, such as 1998. (3) The driving mechanism analysis showed that, overall, precipitation was positively correlated with the water body area and temperature was negatively correlated with the water body area. Additionally, human activities can have an impact on surface water dynamics. The key influencing factors are diverse for each water body type; it was found that seasonal water bodies were correlated with precipitation and paddy fields and permanent water bodies were correlated with temperature and urban construction. The accurate monitoring of the spatial and temporal dynamics of open-surface water performed in this study can shed light on the sustainable development of water resources and the environment.

Estimating the Forest Carbon Storage of Chongming Eco-Island, China, Using Multisource Remotely Sensed Data

Urban forests are highly heterogeneous; information about the combined effect of forest classification scale and algorithm selection on the estimation accuracy for urban forests remains unclear. In this study, we chose Chongming eco-island in the mega-city of Shanghai, a national experimental carbon neutral construction plot in China, as the study object. Remote sensing estimation models (simple regression models vs. machine learning models) of forest carbon density were constructed across different classification scales (all forests, different forest types, and dominant tree species) based on high-resolution aerial photographs and Sentinel-2A remote sensing images, and a large number of field surveys and optimal models were screened by ten-fold cross-validation. The results showed that (1) in early 2020, the total forest area and carbon storage of Chongming eco-island were 307.8 km2 and 573,123.6 t, respectively, among which the areal ratios and total carbon storage ratios of evergreen broad-leaved forest, deciduous broad-leaved forest, and warm coniferous forest were 51.4% and 53.3%, 33.5% and 32.8%, and 15.1% and 13.9%, respectively. (2) The average forest carbon density of Chongming eco-island was 18.6 t/ha, among which no differences were detected among the three forest types (i.e., 17.2–19.2 t/ha), opposite to what was observed among the dominant tree species (i.e., 14.6–23.7 t/ha). (3) Compared to simple regression models, machine learning models showed an improvement in accuracy performance across all three classification scales, with average rRMSE and rBias values decreasing by 29.4% and 53.1%, respectively; compared to the all-forests classification scale, the average rRMSE and rBias across the algorithms decreased by 25.0% and 45.2% at the forest-type classification scale and by 28.6% and 44.3% at the tree species classification scale, respectively. We concluded that refining the forest classification, combined with advanced prediction procedures, could improve the accuracy of carbon storage estimates for urban forests.

Performance Evaluation of Mangrove Species Classification Based on Multi-Source Remote Sensing Data Using Extremely Randomized Trees in Fucheng Town, Leizhou City, Guangdong Province

Mangroves are an important source of blue carbon that grow in coastal areas. The study of mangrove species distribution is the basis of carbon storage research. In this study, we explored the potential of combining optical (Gaofen-1, Sentinel-2, and Landsat-9) and fully polarized synthetic aperture radar data from different periods (Gaofen-3) to distinguish mangrove species in the Fucheng town of Leizhou, Guangdong Province. The Gaofen-1 data were fused with Sentinel-2 and Landsat-9 satellite data, respectively. The new data after fusion had both high spatial and spectral resolution. The backscattering coefficient and polarization decomposition parameters of the fully polarized SAR data which could characterize the canopy structure of mangroves were extracted. Ten different feature combinations were designed by combining the two types of data. The extremely randomized trees algorithm (ERT) was used to classify the species, and the optimal feature subset was selected by the feature selection algorithm on the basis of the ERT, and the importance of the features was sorted. Studies show the following: (1) When controlling a single variable, the higher the spatial resolution of the multi-spectral data, the higher the interspecific classification accuracy. (2) The coupled Sentinel-2 and Landsat-9 data with a 2 m resolution will have higher classification accuracy than a single data source. (3) The selected feature subset contains all types of features in the optical data and the polarization decomposition features of the SAR data from different periods: multi-spectral band &gt; texture feature &gt; polarization decomposition parameter &gt; vegetation index. Among the optimized feature combinations, the classification accuracy of mangrove species was the highest, the overall classification accuracy was 90.13%, and Kappa was 0.84, indicating that multi-source and SAR data from different periods coupling could improve the discrimination of mangrove species. (4) The ERT classification algorithm is suitable for the study of mangrove species classification, and the classification accuracy of extremely random trees in this paper is higher than that of random forest (RF), K-nearest neighbor (KNN), and Bayesian (Bayes). The results can provide technical guidance and data support for mangrove species monitoring based on multi-source satellite data.