Sentinel-2 Images Research Articles

Complementarity of Sentinel-1 and Sentinel-2 Data for Soil Salinity Monitoring to Support Sustainable Agriculture Practices in the Central Bolivian Altiplano

Soil salinization will affect 50% of global cropland areas by 2050 and represents a major threat to agricultural production and food sovereignty. As soil salinity monitoring is costly and time consuming, many regions of the world undertake very limited soil salinity observation (in space and time), preventing the accurate assessment of soil salinity hazards. In this context, this study assesses the relative performance of Sentinel-1 radar and Sentinel-2 optical images, and the combination of the two, for monitoring changes in soil salinity at high spatial and temporal resolution, which is essential to evaluate the mitigation measures required for the sustainable adaptation of agriculture practices. For this purpose, an improved learning database made of 863 soil electrical conductivity (i.e., soil salinity) observations is considered for the training/validation step of a Random Forest (RF) model. The RF model is successively trained with (1) only Sentinel-1, (2) only Sentinel-2 and (3) both Sentinel-1 and -2 features using the Genetic Algorithm (GA) to reduce multi-collinearity in the independent variables. Using k-fold cross validation (3-fold), overall accuracy (OA) values of 0.83, 0.88 and 0.95 are obtained when considering only Sentinel-2, only Sentinel-1 and both Sentinel-1 and -2 features as independent variables. Therefore, these results highlight the clear complementarity of radar (i.e., Sentinel-1) and optical (i.e., Sentinel-2) images to improve soil salinity mapping, with OA increases of approximately 10% and 7% when compared to Sentinel-2 and Sentinel-1 alone. Finally, pre-sowing soil salinity maps over a five-year period (2019–2023) are presented to highlight the benefit of the proposed procedure to support the sustainable management of agricultural lands in the context of soil salinization on a regional scale.

Rice Phenology Classification Model Based on Sentinel-1 Using Machine Learning Method on Google Earth Engine

Rice phenology information is important in supporting planning systems, land management, and making the right decisions to sustainably carry out rice production. This study aimed to determine the best rice phenology classification model by combining VV and VH polarizations on Sentinel-1 images, which produce polarization indices such as the ratio polarization index (RPI), normalized different polarization index (NDPI), and average polarization index (API) using the ensemble random forest (RF) using the Google Earth Engine (GEE) application. This research was conducted in the rice fields of PT Sang Hyang Seri, Subang Regency, West Java. The research data comprised Sentinel-1 SAR GRD satellite imagery data with acquisition modes interferometric wide swath (IW) for 2021–2022 obtained from the GEE platform. In this study, the performance of two machine learning methods for classification was compared: classification and regression trees (CART) and RF. This study found that the best rice phase classification model could be acquired from the RF method with four predictors, namely, API, RPI, NDPI, and slope, with a statistical value of kappa of 98.22%. The RF classification model has better accuracy than the CART classification model.

Spatial Distribution and Variation in Debris Cover and Flow Velocities of Glaciers during 1989–2022 in Tomur Peak Region, Tianshan Mountains

In this study, we utilized a feature optimization method combining texture and topographical factors with the random forest (RF) approach to identify changes in the extent of the debris cover around the Tianshan Tomur Peak between 1989 and 2022. Based on Sentinel-1 image data, we extracted glacier flow velocities using an offset tracking method and conducted a long-term analysis of flow velocities in combination with existing datasets. The debris identification results for 2022 showed that the debris-covered area in the study region was 409.2 km2, constituting 22.8% of the total glacier area. Over 34 years, the area of debris cover expanded by 69.4 km2, reflecting a growth rate of 20.0%. Analysis revealed that glaciers in the Tomur Peak area have been decelerating at an overall rate of −4.0% per decade, with the complexity of the glacier bed environment and the instability of the glacier’s internal structure contributing to significant seasonal and interannual variability in the movement speeds of individual glaciers.

Mapping vertical and horizonal deformation of the newly reclaimed third runway at Hong Kong International Airport with PAZ, COSMO-SkyMed, and Sentinel-1 SAR images

In order to meet long-term air transportation needs, Hong Kong International Airport (HKIA) has been constructing the third runway system (3RS) through land reclamation projects since 2016. However, the third runway and taxiway, which were paved in September 2021, suffered from severe surface settlement. Therefore, we attempted to reveal the settlement pattern within the 3RS and analyze the factors contributing to this displacement. Employing the multi-temporal interferometric synthetic aperture radar (MT-InSAR) technique, we utilized PAZ, COSMO-SkyMed, and Sentinel-1 satellite images to extract ground displacements along the radar line of sight. The results show that, compared with COSMO-SkyMed and Sentinel-1 images, PAZ demonstrates superior performance in the deformation monitoring of the HKIA, with more deformation details and larger displacements. The maximum settlement rate derived by PAZ on the 3RS exceeds 110 mm/year. To ensure the reliability of the derived displacements along the LOS, we validated them through cross-validation, comparing the outcomes from different SAR data with data from GPS stations. In order to gain a detailed deformation information, we derived the vertical and horizontal (ground range direction) displacements by utilizing deformation measurements along the LOS of multiple SAR satellites and the geometric information, and the maximum values recorded for vertical and E-W displacements reached 88.31 mm/y and 41.91 mm/y, respectively. Furthermore, our investigation revealed that the significant local deformation observed on the taxiway and third runway was predominantly attributed to various factors including the creep of soft soil, consolidation of filling materials, characteristics of road surface materials, and distinct stages of construction.

Leafiness-LiDAR index and NDVI for identification of temporal patterns in super-intensive almond orchards as response to different management strategies

The use of super-intensive orchards is a growing trend in fruit production. The present study aims to improve management of these cropping systems by focusing on how agronomic decisions impact orchard dynamics in the short to medium term and by providing a decision-support approach based on stable temporal patterns from previous seasons. A multitemporal study using remote sensing and LiDAR was conducted in a commercial almond orchard over four growing seasons (2019–2022) to determine the optimal timing of image acquisition for variable pre-harvest treatments. A model-based clustering (mclust) was applied to optimal Sentinel-2 NDVI maps and apparent soil electrical conductivity (ECa) data, interpolated to the pixel centroids of Sentinel-2 image grids, to delineate potential management zones (PMZs). The leafiness-LiDAR index (LLI), a leaf area index (LAI) estimator, was obtained as ground truth after summer pruning and before harvesting, showing a significant influence of fertigation and pruning on the LAI, with summer pruning particularly influencing orchard dynamics. The optimal time for NDVI mapping was found to be two months after summer pruning in productive years and two weeks after in unproductive years. The delineated PMZs were consistent across seasons and corresponded to significant LAI differences. This method could contribute to improving resource management and sustainability in super-intensive commercial orchards.

Improving post-fire GEDI canopy height accuracy and canopy height mapping through convolutional neural network model

ABSTRACT The Global Ecosystem Dynamics Investigation (GEDI) LiDAR data have been widely used to measure tree canopy heights. However, due to the altered GEDI waveform geometry in the fire-consumed canopy layers, it is challenging to accurately detect diminished signals with the original GEDI threshold settings, which may lead to underestimation of canopy height under extreme fire conditions. In this study, we delve into the relationship between the accuracy of post-fire GEDI canopy height and front thresholds applied to GEDI waveforms. Notably, thresholds 3σ and 6σ from the GEDI settings proved too high to detect burnt canopy signals, resulting in the lowest accuracy and substantial underestimation. In contrast, the 1σ threshold achieved the highest accuracy. Subsequently, a 1D CNN model was developed to calibrate post-fire GEDI canopy height, utilizing L1B waveforms as input, resulting in a significant reduction of RMSE from 12.2 m to 6.7 m and the elimination of canopy height underestimation in high and extreme fire severity scenarios. In the third experiment, a 2D CNN model was constructed for mapping calibrated canopy height with Sentinel-2 images, revealing a noteworthy drop in RMSE from 14.6 m to 8.5 m compared to the same model using original GEDI canopy height.

Field validation of NDVI to identify crop phenological signatures

Purpose and MethodsCrop identification using remotely sensed imagery provides useful information to make management decisions about land use and crop health. This research used phonecams to acquire the Normalized Difference Vegetation Index (NDVI) of various crops for three crop seasons. NDVI time series from Sentinel (L121-L192) images was also acquired using Google Earth Engine (GEE) for the same period. The resolution of satellite data is low therefore gap filling and smoothening filters were applied to the time series data. The comparison of data from satellite images and phenocam provides useful insight into crop phenology. The results show that NDVI is generally underestimated when compared to phenocam data. The Savitzky-Golay (SG) and some other gap filling and smoothening methods are applied to NDVI time series based on satellite images. The smoothened NDVI curves are statistically compared with daily NDVI series based on phenocam images as a reference.ResultsThe SG method has performed better than other methods like moving average. Furthermore, polynomial order has been found to be the most sensitive parameter in applying SG filter in GEE. Sentinel (L121-L192) image was used to identify wheat during the year 2022–2023 in Sargodha district where experimental fields were located. The Random Forest Machine Leaning algorithm was used in GEE as a classifier.ConclusionThe classification accuracy has been found 97% using this algorithm which suggests its usefulness in applying to other areas with similar agro-climatic characteristics.

Multi-spectral remote sensing bathymetry based on the variation of substrate spectra and empirical linear relationship

The high spatial resolution and wide applicability of multi-spectral remote sensing make it significant for estimating shallow water depth. In areas with mixed substrates, determining the contribution of substrate reflection to the sensor’s received signal can effectively enhance the precision of multi-spectral remote sensing bathymetry. Therefore, this paper proposes a Two Factors constrained Quasi-Analytical Algorithm (TF-QAA). Firstly, the algorithm considers the spectral variation of the mixed substrates. Secondly, a robust correlation between inherent optical parameters and water depth is established in this study, which is used to enforce a linear relationship constraint on the initially estimated water depth. Sentinel-2 image data was used to verify the algorithm in the Antelope Reef area of the South China Sea. The best result from images taken at different times indicates that the TF-QAA method has the average absolute error (MAE) of 0.92 m, the root mean square error (RMSE) of 1.20 m, the average relative error (MRE) of 10.6%, and the correlation coefficient (R2) of 0.97. The results suggest that the algorithm is effective for Case 1 water, providing a novel approach for extensive bathymetric mapping in shallow water areas of islands, with or without LiDAR-measured depth data.

Enhancement of Comparative Assessment Approaches for Synthetic Aperture Radar (SAR) Vegetation Indices for Crop Monitoring and Identification—Khabarovsk Territory (Russia) Case Study

Crop identification at the field level using remote sensing data is a very important task. However, the use of multispectral data for the construction of vegetation indices is sometimes impossible or limited. For such situations, solutions based on the use of time series of synthetic aperture radar (SAR) indices are promising, eliminating the problems associated with cloudiness and providing an assessment of crop development characteristics during the growing season. We evaluated the use of time series of synthetic aperture radar (SAR) indices to characterize crop development during the growing season. The use of SAR imagery for crop identification addresses issues related to cloudiness. Therefore, it is important to choose the SAR index that is the most stable and has the lowest spatial variability throughout the growing season while being comparable to the normalized difference vegetation index (NDVI). The presented work is devoted to the study of these issues. In this study, the spatial variabilities of different SAR indices time series were compared for a single region for the first time to identify the most stable index for use in precision agriculture, including the in-field heterogeneity of crop sites, crop rotation control, mapping, and other tasks in various agricultural areas. Seventeen Sentinel-1B images of the southern part of the Khabarovsk Territory in the Russian Far East at a spatial resolution of 20 m and temporal resolution of 12 days for the period between 14 April 2021 and 1 November 2021 were obtained and processed to generate vertical–horizontal/vertical–vertical polarization (VH/VV), radar vegetation index (RVI), and dual polarimetric radar vegetation index (DpRVI) time series. NDVI time series were constructed from multispectral Sentinel-2 images using a cloud cover mask. The characteristics of time series maximums were calculated for different types of crops: soybean, oat, buckwheat, and timothy grass. The DpRVI index exhibited the highest stability, with coefficients of variation of the time series that were significantly lower than those for RVI and VH/VV. The main characteristics of the SAR and NDVI time series—the maximum values, the dates of the maximum values, and the variability of these indices—were compared. The variabilities of the maximum values and dates of maximum values for DpRVI were lower than for RVI and VH/VV, whereas the variabilities of the maximum values and the dates of maximum values were comparable for DpRVI and NDVI. On the basis of the DpRVI index, classifications were carried out using seven machine learning methods (fine tree, quadratic discriminant, Gaussian naïve Bayes, fine k nearest neighbors or KNN, random under-sampling boosting or RUSBoost, random forest, and support vector machine) for experimental sites covering a total area of 1009.8 ha. The quadratic discriminant method yielded the best results, with a pixel classification accuracy of approximately 82% and a kappa value of 0.67. Overall, 90% of soybean, 74.1% of oat, 68.9% of buckwheat, and 57.6% of timothy grass pixels were correctly classified. At the field level, 94% of the fields included in the test dataset were correctly classified. The paper results show that the DpRVI can be used in cases where the NDVI is limited, allowing for the monitoring of phenological development and crop mapping. The research results can be used in the south of Khabarovsk Territory and in neighboring territories.

Early-Season Crop Mapping by PRISMA Images Using Machine/Deep Learning Approaches: Italy and Iran Test Cases

Despite its high importance for crop yield prediction and monitoring, early-season crop mapping is severely hampered by the absence of timely ground truth. To cope with this issue, this study aims at evaluating the capability of PRISMA hyperspectral satellite images compared with Sentinel-2 multispectral imagery to produce early- and in-season crop maps using consolidated machine and deep learning algorithms. Results show that the accuracy of crop type classification using Sentinel-2 images is meaningfully poor compared with PRISMA (14% in overall accuracy (OA)). The 1D-CNN algorithm, with 89%, 91%, and 92% OA for winter, summer, and perennial cultivations, respectively, shows for the PRISMA images the highest accuracy in the in-season crop mapping and the fastest algorithm that achieves acceptable accuracy (OA 80%) for the winter, summer, and perennial cultivations early-season mapping using PRISMA images. Moreover, the 1D-CNN algorithm shows a limited reduction (6%) in performance, appearing to be the best algorithm for crop mapping within operational use in cross-farm applications. Machine/deep learning classification algorithms applied on the test fields cross-scene demonstrate that PRISMA hyperspectral time series images can provide good results for early- and in-season crop mapping.

Integrating Sentinel 2 Imagery with High-Resolution Elevation Data for Automated Inundation Monitoring in Vegetated Floodplain Wetlands

Monitoring inundation in flow-dependent floodplain wetlands is important for understanding the outcomes of environmental water deliveries that aim to inundate different floodplain wetland vegetation types. The most effective way to monitor inundation across large landscapes is with remote sensing. Spectral water indices are often used to detect water in the landscape, but there are challenges in using them to map inundation within the complex vegetated floodplain wetlands. The current method used for monitoring inundation in the large floodplain wetlands that are targets for environmental water delivery in the New South Wales portion of the Murray–Darling Basin (MDB) in eastern Australia considers the complex mixing of water with vegetation and soil, but it is a time-consuming process focused on individual wetlands. In this study, we developed the automated inundation monitoring (AIM) method to enable an efficient process to map inundation in floodplain wetlands with a focus on the lower Lachlan floodplain utilising 25 Sentinel-2 image dates spanning from 2019 to 2023. A local adaptive thresholding (ATH) approach of a suite of spectral indices combined with best available DEM and a cropping layer were integrated into the AIM method. The resulting AIM maps were validated against high-resolution drone images, and vertical and oblique aerial images. Although instances of omission and commission errors were identified in dense vegetation and narrow creek lines, the AIM method showcased high mapping accuracy with overall accuracy of 0.8 measured. The AIM method could be adapted to other MDB wetlands that would further support the inundation monitoring across the basin.

Spectral enhancement of PlanetScope using Sentinal-2 images to estimate soybean yield and seed composition

Soybean is an essential crop to fight global food insecurity and is of great economic importance around the world. Along with genetic improvements aimed at boosting yield, soybean seed composition also changed. Since conditions during crop growth and development influences nutrient accumulation in soybean seeds, remote sensing offers a unique opportunity to estimate seed traits from the standing crops. Capturing phenological developments that influence seed composition requires frequent satellite observations at higher spatial and spectral resolutions. This study introduces a novel spectral fusion technique called multiheaded kernel-based spectral fusion (MKSF) that combines the higher spatial resolution of PlanetScope (PS) and spectral bands from Sentinel 2 (S2) satellites. The study also focuses on using the additional spectral bands and different statistical machine learning models to estimate seed traits, e.g., protein, oil, sucrose, starch, ash, fiber, and yield. The MKSF was trained using PS and S2 image pairs from different growth stages and predicted the potential VNIR1 (705 nm), VNIR2 (740 nm), VNIR3 (783 nm), SWIR1 (1610 nm), and SWIR2 (2190 nm) bands from the PS images. Our results indicate that VNIR3 prediction performance was the highest followed by VNIR2, VNIR1, SWIR1, and SWIR2. Among the seed traits, sucrose yielded the highest predictive performance with RFR model. Finally, the feature importance analysis revealed the importance of MKSF-generated vegetation indices from fused images.

Assessing the Impact of Heatwaves on Wheat Crop Health in Ludhiana District Using Remote Sensing Technologies

Climate change is becoming one of the major constraints for agricultural production. Impact on the crop yield due to unseasonal rain, hailstorms, and sudden temperature rise is becoming more frequent. The study assesses the damage done by the heat wave in the Ludhiana district of Punjab in the year 2021-22. Satellite datasets have been used named Sentinel-2 and MODIS datasets. By using the Sentinel-2 Images crop discrimination was performed and wheat crop mask was generated for the year 2018-19 to 2021-22. A comparison of the Normalized Difference Vegetation Index (NDVI) was performed for the study period. MODIS dataset was used to calculate the Temperature Condition Index (TCI) and Temperature Stress Index, which reflects the effect of the heat wave that happened in March 2022. APY (Area, Production, Yield) and data on the procurement of wheat for the year 2021-22 and normal years were compared to visualize the impact of heat wave on the various market dynamics such as production, productivity, etc.

Machine-Learning model for estimating sugarcane production at crop level

Yield maps provide essential information for those who manage the field. The anticipated production data will be able to make better decisions on how resources should be used in harvesting, define market strategies and, above all, it will help evaluate treatments used on the crop. Sugar cane is the predominant crop in Huasteca Potosina, Mexico. The proposed Machine Learning model based on Random Forest Regressor integrates time series of vegetation indices extracted from Sentinel-2 images and meteorological data. The R2 and RMSE metrics (0.84 y 0.034) show the effectiveness of the model for prediction.

Effect of Boko Haram Insurgency on Irrigation Activities Using Field and Earth Observation Data in Part of Northeastern Nigeria

The Boko HaramInsurgency in Nigeria has affected the socio-economic activities in most parts of the northeastern part of the country, leading to a low supply of food from agricultural output and increased poverty. Therefore, the main objective of this paper is to analyse the effect of the Boko Haram insurgency on irrigation activities in Kukuwa village. Data was sourced for interviews of 240 respondents and Landsat8_OLI for 2014 and Sentinel 2 for 2015 to 2019. The images were converted to NDVI and used to determine the spectral signatures for each crop using Spectral Matching Techniques (SMTs). The result indicated the majority of the farmers irrigated their farms as > 2 ha (69%) before the crisis, 2 ha (87%) during the crises and > 2 ha (62%) after the crisis, and the majority hired farmers for labour. The intensity of the irrigation was reduced during the crises and it increased after the crises. The result of the SMTs in 2014 indicated a high spectral signature for irrigation activities, with 227.07ha (29.6%) for tomatoes. In 2015 and 2016, low spectral signatures for crops were observed for 541.69ha (69.80%) and 519.5ha (66.94%), respectively. While the spectral result in 2017, 2018, and 2019 indicates a good appreciation of irrigation activities, which shows an increase in most of the dominant crops cultivated in the area, such as tomato, onion, watermelon, and pepper. Among the breakthroughs of this paper is the analysing the reflected signatures from the classified image for different crops through irrigation activities during and after the insurgency in the area. The accuracy assessment indicated fairly good with Landsat but very good with the sentinel images. The study suggests further research into the effect of boko haram insurgency on irrigation activities using field and earth observation data in part of northeastern Nigeria.

Quantifying City- and Street-Scale Urban Tree Phenology from Landsat-8, Sentinel-2, and PlanetScope Images: A Case Study in Downtown Beijing

Understanding the phenology of urban trees can help mitigate the heat island effect by strategically planting and managing trees to provide shade, reduce energy consumption, and improve urban microclimates. In this study, we carried out the first evaluation of high spatial resolution satellite images from Landsat-8, Sentinel-2, and PlanetScope images to quantify urban street tree phenology in downtown Beijing. The major research goals are to evaluate the consistency in pixel-level spring–summer growth period phenology and to investigate the capacity of high-resolution satellite observations to distinguish phenological transition dates of urban street trees. At the city scale, Landsat-8, Sentinel-2, and PlanetScope show similar temporal NDVI trends in general. The pixel-level analysis reveals that green-up date consistency is higher in areas with medium (NDVI > 0.5) to high (NDVI > 0.7) vegetation cover when the impacts of urban surfaces on vegetation reflectance are excluded. Similarly, maturity date consistency significantly increases in densely vegetated pixels with NDVI greater than 0.7. At the street scale, this study emphasizes the efficacy of NDVI time series derived from PlanetScope in quantifying the phenology of common street tree genera, including Poplars (Populus), Ginkgos (Ginkgo), Chinese Scholars (Styphnolobium), and Willows (Salix), in downtown Beijing to improve urban vegetation planning. Based on PlanetScope observations, we found that the four street tree genera have unique phenological patterns. Interestingly, we found that the trees along many major streets, where Chinese Scholars are the major tree genus, have later green-up dates than other areas in downtown Beijing. In conclusion, the three satellite observation datasets prove to be effective in monitoring street tree phenology during the spring–summer growth period in Beijing. PlanetScope is effective in monitoring tree phenology at the street scale; however, Landsat-8 may be affected by the mixture of land covers due to its relatively coarse spatial resolution.

A multi-scale context-aware and batch-independent lightweight network for green tide extraction from SAR images

ABSTRACT The outbreaks of green tide have caused severe harm to the marine environment and human society. Synthetic Aperture Radar (SAR) plays an important role in green tide monitoring by virtue of its high resolution and cloud-free nature. The existing green tide extraction methods still face challenges in identifying multi-scale green tide patches due to noise interference, uneven greyscale and blurred boundaries in SAR images. Meanwhile, the practical application of deep learning methods with high precision is limited due to the complexity of the model and the large amount of computation. Therefore, we propose a multi-scale context-aware and batch-independent lightweight green tide extraction network called MBL-Net. A novel lightweight heterogeneous backbone is designed to extract multi-scale discriminative features and improve segmentation efficiency by using multi-scale selection kernel (MSK) modules and lightweight stages. Meanwhile, Triplet attention module is introduced to improve the internal consistency of the green tide region and suppress the effect of speckle noise. Then, the mixed pooling-based channel prior module (MCPM) is used to expand the receptive field of the network and extract the fine green tide structure by fusing multi-scale features. In addition, Filter Response Normalisation (FRN) is innovatively applied for feature normalization in the decoding stage, eliminating batch dependency. In order to verify the effectiveness of the proposed method, a dataset is built using the Sentinel-1 images of the Yellow Sea, China, from 2019 to 2021. The experimental results show that the proposed method achieves an overall accuracy of 98.59% with 0.970 G FLOPs and 3.525 M parameters, which ensures high precision and improves green tide detection efficiency. Compared with several representative networks, this method can capture more details of green tide with fewer parameters and faster calculation speed.

EWMACD Algorithm in Early Detection of Defoliation Caused by Dendrolimus tabulaeformis Tsai et Liu

The persistent increase in forest pest outbreaks requires timely detection methods to monitor the disaster precisely. However, early detection is challenging due to insufficient temporal observation and subtle tree changes. This article proposed a novel framework that collaborates multi-source remote sensing data and uses a change detection algorithm to archive early detection of infestation caused by Dendrolimus tabulaeformis Tsai et Liu (D. tabulaeformis) attacks. First, all available Sentinel-2 images with less than 20% cloud cover were utilized. During periods with long intervals (>16 days) between Sentinel-2 images, Landsat-8 images with less than 20% cloud cover were downscaled to a spatial resolution of 10 m using a deep learning algorithm to meet the requirement for a high temporal frequency of clear observations. Second, the spectral index differences between healthy and infested trees were examined to address the challenge of detecting subtle changes in pest attacks. The Enhanced Vegetation Index (EVI) was selected for early defoliation detection. On this basis, the EWMACD (Exponentially Weighted Moving Average Change Detection) algorithm, which is sensitive to subtle changes, was enhanced to improve the capability of detecting early D. tabulaeformis attacks. The assessment showed that the overall accuracy of the change detection (F1 score) reached 0.86 during the early stage and 0.88 during the late stage. The temporal accuracy (Precision) was 84.1% during the early stage. The accuracy significantly improved compared to using a single remote sensing data source. This study presents a new framework capable of monitoring early forest defoliation caused by D. tabulaeformis attacks and offering opportunities for predicting future outbreaks and implementing preventive measures.

Matching Spring Phenology Indicators in Ground Observations and Remote-Sensing Metrics

Accurate monitoring of leaf phenology, from individual trees to entire ecosystems, is vital for understanding and modeling forest carbon and water cycles, as well as assessing climate change impact. However, the accuracy of many remote-sensing phenological products remains difficult to directly corroborate using ground-based monitoring, owing to variations in the observed indicators and the scales used. This limitation hampers the practical implementation of remote-sensing phenological metrics. In our study, the start of growing season (SOS) from 2016 to 2021 was estimated for the continental USA using Sentinel-2 images. The results were then matched with several ground-based spring vegetation phenology metrics obtained by the USA National Phenology Network (USA-NPN). In this study, we focused on the relationships between the leaf-unfolding degree (LUD), the SOS, and the factors that drive these measures. Our results revealed that: (1) the ground-based leaves and increasing leaf size stages were significantly correlated with the SOS; (2) with the closest match being observed for a leaf spread of 13%; (2) the relationship between the SOS and LUD varied according to the species and ecoregion, and the pre-season cumulative radiation was found to be the main factor affecting the degree of matching between the ground observations and the metrics derived from the Sentinel-2 data. Our investigations provide a ground-based spring phenology metric that can be used to verify or evaluate remote-sensing spring phenology products and will help to improve the accuracy of remote-sensing phenology metrics.

Wind speed retrieval approach for VV-polarized synthetic aperture radar during tropical cyclone based on XGBoost model

ABSTRACT In this study, a machine learning approach is applied to wind speed retrieval from vertical-vertical (VV) polarized synthetic aperture radar (SAR) during tropical cyclones (TC). More than 2400 dual-polarized (VV and vertical-horizontal (VH)) Sentinel-1 (S-1) images acquired in interferometric wide (IW) and extra wide (EW) mode are collected in 2016–2022. Along-track observations from stepped-frequency microwave radiometer (SFMR) and soil moisture active passive (SMAP) are available for several images. The azimuthal cut-off wavelength (ACW) is derived from VV-polarized image and the Doppler centroid anomaly (DCA) is provided from Ocean (OCN) product. It is found that SFMR wind speed linearly relates with ACW and DCA. Following this rationale, a machine learning method, called eXtreme Gradient Boosting (XGBoost), is trained for constructing SAR wind speed retrieval geophysical model function (GMF) through abundant matchups from 2000 images collocated with wind speeds from soil moisture active passive (SMAP) microwave radiometer and SFMR, denoted as TCWIND2-S1. Then GMF TCWIND2-S1 is applied for more than 400 images and the validation against SMAP and SFMR products up to 70 m s−1 shows a 3.01 m s−1 root mean squared error (RMSE) with a 0.16 scatter index (SI) and a 0.92 correlation (COR). Retrieval is also compared with CyclObs wind products derived from more than 400 dual-polarized images, yielding a 2.66 m s−1 RMSE with a 0.19 SI and a 0.93 COR. Our work provides an approach for VV-polarized SAR wind retrieval at wind speed ranged from 0 to 70 m s−1 during TCs. This work is conveniently adopted for TerraSAR-X without operating at cross polarization channel.