Dense Time Series Research Articles

Exploring the Relationship Between Time Series of Sentinel-1 Interferometric Coherence Data and Wild Edible Mushroom Yields in Mediterranean Forests

Edible wild mushrooms constitute a valuable marketable non-wood forest product with high relevance worldwide. There is growing interest in developing tools for estimation of mushroom yields and to evaluate the effects that global change may have on them. Remote sensing is a powerful technology for characterization of forest structure and condition, both essential factors in triggering mushroom production, together with meteo-climatic factors. In this work, we explore options to apply synthetic aperture radar (SAR) data from C-band Sentinel-1 to characterize, at the plot level, wild mushroom productive forests in the Mediterranean region, which provide saprotroph and ectomycorrhizal mushrooms. Seventeen permanent plots with mushroom yield data collected weekly during the productive season are characterized with dense time series of Sentinel-1 backscatter intensity (VV and VH polarizations) and 6-day interval interferometric VV coherence during the 2018–2021 period. Weekly-regularized series of SAR data are decomposed with a LOESS approach into trend, seasonality, and remainder. Trends are explored with the Theil-Sen test, and periodicity is characterized by the Discrete Fast Fourier transform. Seasonal patterns of SAR time-series are described and related to mycorrhizal and saprotroph guilds separately. Our results indicate that time series of interferometric coherence show cyclic patterns which might be related with annual mushroom yields and may constitute an indicator of triggering factors in mushroom production, whereas backscatter intensity is strongly correlated with precipitation, making noisy signals without a clear interpretable pattern. Exploring the potential of remotely sensed data for prediction and quantification of mushroom yields contributes to improve our understanding of fungal biological cycles and opens new ways to develop tools that improve its sustainable, efficient, and effective management.

Analyzing Temporal Characteristics of Winter Catch Crops Using Sentinel-1 Time Series

Catch crops are intermediate crops sown between two main crop cycles. Their adoption into the cropping system has increased considerably in the last years due to its numerous benefits, in particular its potential in carbon fixation and preventing nitrogen leaching during winter. The growth period of catch crops in Germany is often marked by dense cloud cover, which limits land surface monitoring through optical remote sensing. In such conditions, synthetic aperture radar (SAR) emerges as a viable option. Despite the known advantages of SAR, the understanding of temporal behavior of radar parameters in relation to catch crops remains largely unexplored. Hence, in this study, we exploited the dense time series of Sentinel-1 data within the Copernicus Space Component to study the temporal characteristics of catch crops over a test site in the center of Germany. Radar parameters such as VV, VH, VH/VV backscatter, dpRVI (dual-pol Radar Vegetation Index) and VV coherence were extracted, and temporal profiles were interpreted for catch crops and preceding main crops along with in situ, temperature, and precipitation data. Additionally, we examined the temporal profiles of winter main crops (winter oilseed rape and winter cereals), that are grown parallel to the catch crop growing cycle. Based on the analyzed temporal patterns, we defined 22 descriptive features from VV, VH, VH/VV and dpRVI, which are specific to catch crop identification. Then, we conducted a Kruskal–Wallis test on the extracted parameters, both crop-wise and group-wise, to assess the significance of statistical differences among different catch crop groups. Our results reveal that there exists a unique temporal pattern for catch crops compared to main crops, and each of these extracted parameters possess a different sensitivity to catch crops. Parameters VV and VH are sensitive to phenological stages and crop structure. On the other hand, VH/VV and dpRVI were found to be highly sensitive to crop biomass. Coherence can be used to detect the sowing and harvest events. The preceding main crop analysis reveals that winter wheat and winter barley are the two dominant main crops grown before catch crops. Moreover, winter main crops (winter oilseed rape, winter cereals) cultivated during the catch crop cycle can be distinguished by exploiting the observed sowing window differences. The extracted descriptive features provide information about sowing, harvest, vigor, biomass, and early/late die-off nature specific to catch crop types. In the Kruskal–Wallis test, the observed high H-statistic and low p-value in several predictors indicates significant variability at 0.001 level. Furthermore, Dunn’s post hoc test among catch crop group pairs highlights the substantial differences between cold-sensitive and legume groups (p < 0.001).

The phenology and water level time-series mangrove index for improved mangrove monitoring

Mangroves face decline and degradation due to human activities and natural forces, making their accurate mapping and dynamic monitoring essential. However, most of the existing mangrove indices that rely on multispectral image spectral characteristics suffer from limitations in terms of recognition accuracy and universality. Therefore, this study aimed to develop a robust and efficient Phenology and Water level Time-series Mangrove Index (PWTMI) for mangrove monitoring. PWTMI is constructed by combining spectral and temporal characteristics from dense time-series multispectral data, wherein phenology and water level time-series characteristics are extracted from NDVI and MNDWI time series. The results show that PWTMI outperforms existing multispectral-based mangrove indices and has an accuracy similar to a hyperspectral-based mangrove index, with overall accuracy ranging from 91.49% to 98.83% and F1 score ranging from 0.91 to 0.98 in four typical areas in China, indicating great potential for long time-series and large-scale mangrove monitoring.

STSNet: A cross-spatial resolution multi-modal remote sensing deep fusion network for high resolution land-cover segmentation

Recently, deep learning models have found extensive application in high-resolution land-cover segmentation research. However, the most current research still suffers from issues such as insufficient utilization of multi-modal information, which limits further improvement in high-resolution land-cover segmentation accuracy. Moreover, differences in the size and spatial resolution of multi-modal datasets collectively pose challenges to multi-modal land-cover segmentation. Therefore, we propose a high-resolution land-cover segmentation network (STSNet) with cross-spatial resolution spatio-temporal-spectral deep fusion. This network effectively utilizes spatio-temporal-spectral features to achieve information complementary among multi-modal data. Specifically, STSNet consists of four components: (1) A high resolution and multi-scale spatial-spectral encoder to jointly extract subtle spatial-spectral features in hyperspectral and high spatial resolution images. (2) A long-term spatio-temporal encoder formulated by spectral convolution and spatio-temporal transformer block to simultaneously delineates the spatial, temporal and spectral information in dense time series Sentinel-2 imagery. (3) A cross-resolution fusion module to alleviate the spatial resolution differences between multi-modal data and effectively leverages complementary spatio-temporal-spectral information. (4) A multi-scale decoder integrates multi-scale information from multi-modal data. We utilized airborne hyperspectral remote sensing imagery from the Shenyang region of China in 2020, with a spatial resolution of 1authors declare that they have no known competm, a spectral number of 249, and a spectral resolution ≤ 5 nm, and its Sentinel dense time-series images acquired in the same period with a spatial resolution of 10 m, a spectral number of 10, and a time-series number of 31. These datasets were combined to generate a multi-modal dataset called WHU-H2SR-MT, which is the first open accessed large-scale high spatio-temporal-spectral satellite remote sensing dataset (i.e., with >2500 image pairs sized 300 m × 300 m for each). Additionally, we employed two open-source datasets to validate the effectiveness of the proposed modules. Extensive experiments show that our multi-scale spatial-spectral encoder, spatio-temporal encoder, and cross-resolution fusion module outperform existing state-of-the-art (SOTA) algorithms in terms of overall performance on high-resolution land-cover segmentation. The new multi-modal dataset will be made available at http://irsip.whu.edu.cn/resources/resources_en_v2.php, along with the corresponding code for accessing and utilizing the dataset at https://github.com/RS-Mage/STSNet.

Forest disturbance regimes and trends in continental Spain (1985–2023) using dense landsat time series

Forest disturbance regimes across biomes are being altered by interactive effects of global change. Establishing baselines for assessing change requires detailed quantitative data on past disturbance events, but such data are scarce and difficult to obtain over large spatial and temporal scales. The integration of remote sensing with dense time series analysis and cloud computing platforms is enhancing the ability to monitor historical disturbances, and especially non-stand replacing events along climatic gradients. Since the integration of such tools is still scarce in Mediterranean regions, here, we combine dense Landsat time series and the Continuous Change Detection and Classification - Spectral Mixture Analysis (CCDC-SMA) method to monitor forest disturbance in continental Spain from 1985 to 2023. We adapted the CCDC-SMA method for improved disturbance detection creating new spectral libraries representative of the study region, and quantified the year, month, severity, return interval, and type of disturbance (stand replacing, non-stand replacing) at a 30 m resolution. In addition, we characterised forest disturbance regimes and trends (patch size and severity, and frequency of events) of events larger than 0.5 ha at the national scale by biome (Mediterranean and temperate) and forest type (broadleaf, needleleaf and mixed). We quantified more than 2.9 million patches of disturbed forest, covering 4.6 Mha over the region and period studied. Forest disturbances were on average larger but less severe in the Mediterranean than in the temperate biome, and significantly larger and more severe in needleleaf than in mixed and broadleaf forests. Since the late 1980s, forest disturbances have decreased in size and severity while increasing in frequency across all biomes and forest types. These results have important implications as they confirm that disturbance regimes in continental Spain are changing and should therefore be considered in forest strategic planning for policy development and implementation.

Global Landslide Finder: Detecting the Time and Place of Landslides with Dense Earth Observation Time Series

This paper presents a remote sensing approach for rapidly and automatically generating maps of surface disturbances caused by landslides on the global scale. Our approach not only identifies the locations of these disturbances but also pinpoints the estimated time of their occurrence. Using the Continuous Change Detection and Classification (CCDC) algorithm within the Google Earth Engine (GEE) platform, we analyzed two decades of Landsat 5, 7, and 8 surface reflectance data. We tested this approach in five landslide-prone regions: Iburi (Japan), Kashmir (Pakistan), Karnataka (India), Porgera (Papua New Guinea), and Pasang Lhamu (Nepal). The results were promising, with R2 values ranging up to 0.85, indicating a robust correlation between detected disturbances and actual landslide events compared to manually made inventories. The accuracy metrics further validated our method, with a producer’s accuracy of 75%, a user’s accuracy of 73%, and an F1 score of 75%. Furthermore, the method proved well transferable across different locations. These findings demonstrate the method’s potential as a valuable tool for near real-time and historical analysis of landslide activity, thereby contributing to global disaster management and mitigation efforts.

Harmonized Landsat and Sentinel-2 Data with Google Earth Engine

Continuous and dense time series of satellite remote sensing data are needed for several land monitoring applications, including vegetation phenology, in-season crop assessments, and improving land use and land cover classification. Supporting such applications at medium to high spatial resolution may be challenging with a single optical satellite sensor, as the frequency of good-quality observations can be low. To optimize good-quality data availability, some studies propose harmonized databases. This work aims at developing an ‘all-in-one’ Google Earth Engine (GEE) web-based workflow to produce harmonized surface reflectance data from Landsat-7 (L7) ETM+, Landsat-8 (L8) OLI, and Sentinel-2 (S2) MSI top of atmosphere (TOA) reflectance data. Six major processing steps to generate a new source of near-daily Harmonized Landsat and Sentinel (HLS) reflectance observations at 30 m spatial resolution are proposed and described: band adjustment, atmospheric correction, cloud and cloud shadow masking, view and illumination angle adjustment, co-registration, and reprojection and resampling. The HLS is applied to six equivalent spectral bands, resulting in a surface nadir BRDF-adjusted reflectance (NBAR) time series gridded to a common pixel resolution, map projection, and spatial extent. The spectrally corresponding bands and derived Normalized Difference Vegetation Index (NDVI) were compared, and their sensor differences were quantified by regression analyses. Examples of HLS time series are presented for two potential applications: agricultural and forest phenology. The HLS product is also validated against ground measurements of NDVI, achieving very similar temporal trajectories and magnitude of values (R2 = 0.98). The workflow and script presented in this work may be useful for the scientific community aiming at taking advantage of multi-sensor harmonized time series of optical data.

Reconstruction of dense time series high spatial resolution NDVI data using a spatiotemporal optimal weighted combination estimation model based on Sentinel-2 and MODIS

A regular, dense time series of the normalized difference vegetation index (NDVI) is a crucial remote sensing parameter that provides insights into the growth status of plants across different temporal and spatial scales. However, the production of high-resolution, equal-interval, and dense-time-series NDVI products currently faces technical challenges, limiting their application in agriculture and forestry-related fields. This study proposes a new method for reconstructing a dense Sentinel-2 NDVI time series based on the spatiotemporal optimal weighted combination estimation model (SOWCEM). The SOWCEM is developed using state-of-the-art spatial and temporal reconstruction algorithms. This model considers the distribution characteristics of errors in spatiotemporal dimensions, allowing for adaptive determination of the optimal combination weight at the pixel level. Then, the reconstructed regular dense-time-series NDVI data covering the region for five days/time are evaluated through a comparative analysis. The experimental results demonstrate that the regional NDVI reconstruction image based on the SOWCEM algorithm has a determination coefficient (R2) of 0.9082, a root mean square error (RMSE) of 0.0403, and a comprehensive overall error (ERGAS) of 5.7726. Compared with single spatiotemporal dimension models, the SOWCEM algorithm shows an average improvement of 5.78% in R2, while the RMSE and ERGAS decrease by 6.5% and 6.86%, respectively. Moreover, the stability and robustness of the models are also enhanced. These results indicate that the SOWCEM algorithm can effectively achieve a high-quality fusion reconstruction of regional images, significantly enhancing the accuracy of high-spatial-resolution NDVI dense temporal spectral reconstruction. Furthermore, it exhibits clear superiority and good applicability. This algorithmic framework provides a feasible approach for the dense reconstruction of high-spatial-resolution regular time-series NDVI data, enabling more accurate and efficient high-resolution NDVI remote sensing monitoring services in related field applications. The core code of SOWCEM is available at https://github.com/GISerZkun/SOWCEM.

Comparison of anomalous displacement detectability of bridges between X and C band InSAR data: Case study on a collapse of water pipe bridge

This paper presents a feasibility of detecting anomalous displacements on bridges where collapses have occurred using C-band satellite synthetic aperture radar (SAR). Recently, the number of aging bridges due to damage and corrosion has been increased in many countries. Although there are various maintenance and management systems, the authors have focused on satellite SAR, one of the remote sensing technologies, to establish a wide-area multiple bridge monitoring technology. In a previous study, the authors analyzed high-resolution X-band interferometric SAR (InSAR) data acquired by the Italian satellite COSMO-SkyMed during the two years before the accident for the MUSOTA water pipe bridge in Japan, which collapsed on October 2021. As a result, it was shown that there was an anomalous displacement about one year before the accident, which seemed to be a sign of the collapse. On the other hand, the utilization of Sentinel-1 (C-band) data, which has lower resolution than X-band data, but is freely available, covers a wide area, and has dense time-series SAR images, would be beneficial for the practical application of this technology. In this study, InSAR data acquired by Sentinel-1 for the MUSOTA water pipe bridge from 2016 to 2022 are analyzed. Then, we compare the line-of-sight (LOS) displacement anomaly before the collapse with the results analyzed by the X-band satellite. As a result of statistical analysis of the displacement between the collapsed span and the adjacent spans with similar structural type assumed to be intact, it is shown that Sentinel-1 also captured differences which seemed to be the signs of the collapse about one year before the accident. This result supports the analysis of the X-band satellite data and shows potential for utilization in bridge anomaly detection systems, even for low-resolution C-band data.

An Improved Gap-Filling Method for Reconstructing Dense Time-Series Images from LANDSAT 7 SLC-Off Data

Over recent decades, Landsat satellite data has evolved into a highly valuable resource across diverse fields. Long-term satellite data records with integrity and consistency, such as the Landsat series, provide indispensable data for many applications. However, the malfunction of the Scan Line Corrector (SLC) on the Landsat 7 satellite in 2003 resulted in stripping in subsequent images, compromising the temporal consistency and data quality of Landsat time-series data. While various methods have been proposed to improve the quality of Landsat 7 SLC-off data, existing gap-filling methods fail to enhance the temporal resolution of reconstructed images, and spatiotemporal fusion methods encounter challenges in managing large-scale datasets. Therefore, we propose a method for reconstructing dense time series from SLC-off data. This method utilizes the Neighborhood Similar Pixel Interpolator to fill in missing values and leverages the time-series information to reconstruct high-resolution images. Taking the blue band as an example, the surface reflectance verification results show that the Mean Absolute Error (MAE) and BIAS reach minimum values of 0.0069 and 0.0014, respectively, with the Correlation Coefficient (CC) and Structural Similarity Index Metric (SSIM) reaching 0.93 and 0.94. The proposed method exhibits advantages in repairing SLC-off data and reconstructing dense time-series data, enabling enhanced remote sensing applications and reliable Earth’s surface reflectance data reconstruction.

Land Cover Classification Based on Multimodal Remote Sensing Fusion

Abstract. Global high-precision and high timeliness land cover data is a fundamental and strategic resource for global strategic interest maintenance, global environmental change research, and sustainable development planning. However, due to difficulties in obtaining control and reference information from overseas, a single data source cannot effectively cover, and surface coverage classification faces significant challenges in information extraction. Based on this, this article proposes an intelligent interpretation method for typical elements based on multimodal fusion, starting from the characteristics of domestic remote sensing images. It also develops an optical SAR data conversion and complementarity strategy based on convolutional translation networks, as well as a typical element extraction algorithm. This solves the problems of sparse remote sensing images, limited effective observations, and difficult information recognition, thereby achieving automation of typical element information under dense observation time series High precision extraction and analysis.

Sensitivity of Sentinel-1 Backscatter to Management-Related Disturbances in Temperate Forests

The rapid and accurate detection of forest disturbances in temperate forests has become increasingly crucial as policy demands and climate pressure on these forests rise. The cloud-penetrating Sentinel-1 radar constellation provides frequent and high-resolution observations with global coverage, but few studies have assessed its potential for mapping disturbances in temperate forests. This study investigated the sensitivity of temporally dense C-band backscatter data from Sentinel-1 to varying management-related disturbance intensities in temperate forests, and the influence of confounding factors such as radar backscatter signal seasonality, shadow, and layover on the radar backscatter signal at a pixel level. A unique network of 14 experimental sites in the Netherlands was used in which trees were removed to simulate different levels of management-related forest disturbances across a range of representative temperate forest species. Results from six years (2016–2022) of Sentinel-1 observations indicated that backscatter seasonality is dependent on species phenology and degree of canopy cover. The backscatter change magnitude was sensitive to medium- and high-severity disturbances, with radar layover having a stronger impact on the backscatter disturbance signal than radar shadow. Combining ascending and descending orbits and complementing polarizations compared to a single orbit or polarization was found to result in a 34% mean increase in disturbance detection sensitivity across all disturbance severities. This study underlines the importance of linking high-quality experimental ground-based data to dense satellite time series to improve future forest disturbance mapping. It suggests a key role for C-band backscatter time series in the rapid and accurate large-area monitoring of temperate forests and, in particular, the disturbances imposed by logging practices or tree mortality driven by climate change factors.

Temporally transferable crop mapping with temporal encoding and deep learning augmentations

Detailed maps on the spatial and temporal distribution of crops are key for a better understanding of agricultural practices and for food security management. Multi-temporal remote sensing data and deep learning (DL) have been extensively studied for deriving accurate crop maps. However, strategies to solve the problem of transferring crop classification models over time, e.g., training the model with data for a recent year and mapping back to the past, have not been fully explored. This is due to the lack of a generalized method for aggregating optical data with regard to the irregularity in annual clear sky observations and the scarcity of multi-annual crop reference data to support a more generalized DL model. In this study, we tackled these challenges by introducing a method namely Temporal Encoding (TE) to capture the irregular phenological information. Subsequently, we adapted and integrated two methods, i.e., Random Observations Selection (ROS) and Random Day Shifting (RDS) to simulate the variability of temporal sparsity as well as the shifts of crop phenology over different years. We tested this approach with a 1-dimensional Convolutional Neural Network (1D-CNN) and a Transformer Network models. Our results for both classifiers showed that models trained with crop reference data from 2018 and a dense time series of Landsat 7/8 and Sentinel-2 A/B data can be transferred with little decreases in accuracy to map 12 consecutive years from 2010 to 2021. The Transformer Network was slightly more accurate, while the 1D-CNN was much three times faster. Furthermore, the proposed models could achieve similar performances in the same years with and without fully available satellite information. The TE with ROS and RDS appears well suited for improving temporal transferability to support long term historic crop mapping.

Historical information fusion of dense multi-source satellite image time series for flood extent mapping

Existing lightweight flood extent mapping (FEM) methods, leveraging satellite imagery, offer respite from the burden of extensive data processing. However, these methods often grapple with suboptimal accuracy of the FEM results. To tackle this challenge, this paper introduces an innovative lightweight historical information fusion framework aimed at enhancing FEM results through the utilization of dense multi-source satellite image time series (MSSITS). Firstly, we propose an ultra-lightweight convolutional neural network comprising fundamental network modules, such as convolution layers, batch normalization layers, and parametric rectified linear unit layers. This network swiftly extracts initial FEM results from dense MSSITS. Secondly, we devise a historical information fusion strategy to refine the initial FEM results of the current image by fusing historical FEM results exhibiting high temporal correlation. During the fusion stage, we harness the complementary attributes of dense MSSITS. Specifically, the fusion of diverse satellite sensor data aids in mitigating false alarms in the initial FEM results, thereby facilitating the acquisition of refined FEM results. This approach effectively diminishes the impact of noise interferences, including terrain shadows and surging flood waves in synthetic aperture radar images, as well as cloud occlusions and cloud shadows in optical images. Experimental findings, based on the dense MSSITS of four authentic flood events, demonstrate that our proposed method yields relatively superior FEM performance while maintaining a low FEM complexity in comparison with commonly used semantic segmentation methods.

Dense Time Series Generation of Surface Water Extents through Optical–SAR Sensor Fusion and Gap Filling

Surface water is a vital component of the Earth’s water cycle and characterizing its dynamics is essential for understanding and managing our water resources. Satellite-based remote sensing has been used to monitor surface water dynamics, but cloud cover can obscure surface observations, particularly during flood events, hindering water identification. The fusion of optical and synthetic aperture radar (SAR) data leverages the advantages of both sensors to provide accurate surface water maps while increasing the temporal density of unobstructed observations for monitoring surface water spatial dynamics. This paper presents a method for generating dense time series of surface water observations using optical–SAR sensor fusion and gap filling. We applied this method to data from the Copernicus Sentinel-1 and Landsat 8 satellite data from 2019 over six regions spanning different ecological and climatological conditions. We validated the resulting surface water maps using an independent, hand-labeled dataset and found an overall accuracy of 0.9025, with an accuracy range of 0.8656–0.9212 between the different regions. The validation showed an overall false alarm ratio (FAR) of 0.0631, a probability of detection (POD) of 0.8394, and a critical success index (CSI) of 0.8073, indicating that the method generally performs well at identifying water areas. However, it slightly underpredicts water areas with more false negatives. We found that fusing optical and SAR data for surface water mapping increased, on average, the number of observations for the regions and months validated in 2019 from 11.46 for optical and 55.35 for SAR to 64.90 using both, a 466% and 17% increase, respectively. The results show that the method can effectively fill in gaps in optical data caused by cloud cover and produce a dense time series of surface water maps. The method has the potential to improve the monitoring of surface water dynamics and support sustainable water management.

Ecological restoration trajectory of the Taitema Lake wetland in arid northwest China: A 36-year wetland health assessment using Landsat time series data

The Taitema Lake wetland in the lower reaches of the Tarim River (northwest China) is sensitive to hydrological changes and provides necessary ecosystem functions for biodiversity conservation. Monitoring and evaluating the long-term dynamics of the Taitema Lake wetland is essential for conserving and restoring regional wetland ecosystems. This paper used a dense time series of Landsat images from 1986 to 2021 to map the Taitema Lake wetlands. It analyzed the annual and seasonal variations of the wetlands by ten ecological indicators, which include area extent, vegetation, hydrology, and landscape features. A systematic assessment of the ecological quality of the wetlands over the past 36 years, along with their influencing factors, was conducted from the perspective of wetland health. The results showed that (1) the recovered water areas and the wetlands showed a clear trend of ecological restoration since 2000, accompanied by an increased fragmentation. The total wetland area increased from 69.95 km2 in 1986 to 1164.47 km2 in 2021, with a 15.6-fold increase, and the vegetation area increased twentyfold to 639.84 km2. (2) The wetness in the core zone increased consistently across seasons, coupled with a simultaneous vegetation expansion surrounding the core zone. The areas with increasing wetness levels account for 48 % during winter and spring, and the areas with vegetation expansion account for 81 % and 72 % in summer and autumn, respectively. (3) The wetland ecological health status recovered to a “Good” level in 2021 and has not reached the “Excellent” status. The expansion of the water body is the primary driver behind the promotion of wetland ecological recovery. Small water bodies within the area range of 0.05 to 0.1 km2 notably exhibited the most stimulating effects on the wetlands. Thus, increasing the count of smaller water bodies helps rehabilitate vegetation growth and mitigating the water supply’s stress to sustain large lacustrine bodies.

Evaluation and Selection of Multi-Spectral Indices to Classify Vegetation Using Multivariate Functional Principal Component Analysis

The identification, classification and mapping of different plant communities and habitats is of fundamental importance for defining biodiversity monitoring and conservation strategies. Today, the availability of high temporal, spatial and spectral data from remote sensing platforms provides dense time series over different spectral bands. In the case of supervised mapping, time series based on classical vegetation indices (e.g., NDVI, GNDVI, …) are usually input characteristics, but the selection of the best index or set of indices (which guarantees the best performance) is still based on human experience and is also influenced by the study area. In this work, several different time series, based on Sentinel-2 images, were created exploring new combinations of bands that extend the classic basic formulas as the normalized difference index. Multivariate Functional Principal Component Analysis (MFPCA) was used to contemporarily decompose the multiple time series. The principal multivariate seasonal spectral variations identified (MFPCA scores) were classified by using a Random Forest (RF) model. The MFPCA and RF classifications were nested into a forward selection strategy to identify the proper and minimum set of indices’ (dense) time series that produced the most accurate supervised classification of plant communities and habitat. The results we obtained can be summarized as follows: (i) the selection of the best set of time series is specific to the study area and the habitats involved; (ii) well-known and widely used indices such as the NDVI are not selected as the indices with the best performance; instead, time series based on original indices (in terms of formula or combination of bands) or underused indices (such as those derivable with the visible bands) are selected; (iii) MFPCA efficiently reduces the dimensionality of the data (multiple dense time series) providing ecologically interpretable results representing an important tool for habitat modelling outperforming conventional approaches that consider only discrete time series.

Vegetation Classification and Evaluation of Yancheng Coastal Wetlands Based on Random Forest Algorithm from Sentinel-2 Images

The identification of wetland vegetation is essential for environmental protection and management as well as for monitoring wetlands’ health and assessing ecosystem services. However, some limitations on vegetation classification may be related to remote sensing technology, confusion between plant species, and challenges related to inadequate data accuracy. In this paper, vegetation classification in the Yancheng Coastal Wetlands is studied and evaluated from Sentinel-2 images based on a random forest algorithm. Based on consistent time series from remote sensing observations, the characteristic patterns of the Yancheng Coastal Wetlands were better captured. Firstly, the spectral features, vegetation indices, and phenological characteristics were extracted from remote sensing images, and classification products were obtained by constructing a dense time series using a dataset based on Sentinel-2 images in Google Earth Engine (GEE). Then, remote sensing classification products based on the random forest machine learning algorithm were obtained, with an overall accuracy of 95.64% and kappa coefficient of 0.94. Four indicators (POP, SOS, NDVIre, and B12) were the main contributors to the importance of the weight analysis for all features. Comparative experiments were conducted with different classification features. The results show that the method proposed in this paper has better classification.

Tracking effects of extreme drought on coniferous forests from space using dynamic habitat indices

Terrestrial ecosystems such as coniferous forests in Central Europe are experiencing changes in health status following extreme droughts compounding with severe heat waves. The increasing temporal resolution and spatial coverage of earth observation data offer new opportunities to assess these dynamics. Dense time-series of optical satellite data allow for computing Dynamic Habitat Indices (DHIs), which have been predominantly used in biodiversity studies. However, DHIs cover three aspects of vegetation changes that could be affected by drought: annual productivity, minimum cover, and seasonality. Here, we evaluate the health status of coniferous forests in the federal state of Hesse in Germany over the period 2017–2020 including the severe drought year of 2018 using DHIs based on the Normalized Difference Vegetation Index (NDVI) for drought assessment. To identify the most important variables affecting coniferous forest die-off, a series of environmental variables together with the three DHIs components were used in a logistic regression (LR) model. Each DHI component changed significantly across non-damaged and damaged sites in all years (p-value 0.05). When comparing 2017 to 2019, DHI-based annual productivity decreased and seasonality increased. Most importantly, none of the DHI components had reached pre-drought conditions, which likely indicates a change in ecosystem functioning. We also identified spatially explicit areas highly affected by drought. The LR model revealed that in addition to common environmental parameters related to temperature, precipitation, and elevation, DHI components were the most important factors explaining the health status. Our analysis demonstrates the potential of DHIs to capture the effect of drought events on Central European coniferous forest ecosystems. Since the spaceborne data are available at the global level, this approach can be applied to track the dynamics of ecosystem conditions in other regions, at larger spatial scales, and for other Land Use/Land Cover types.

Tracking the dynamics of tidal wetlands with time-series satellite images in the Yangtze River Estuary, China

ABSTRACT Tidal wetlands provide a variety of ecosystem services to coastal communities but suffer severe losses due to anthropogenic activities in the Yangtze River Estuary (YRE). However, the detailed dynamics of tidal wetlands have not been well studied with sufficient spatiotemporal resolution. Here, we proposed a rapid classification method that integrates the COntinuous monitoring of Land Disturbance (COLD) algorithm and Median Composite (MC) based on the dense Landsat time series to track the dynamic processes of tidal wetlands in the YRE from 1990 to 2020. The results showed that the COLD-MC demonstrated remarkable effectiveness in detecting the change of tidal wetlands and excellent overall accuracy and kappa coefficient ranging from 90% to 96% and 0.89–0.95, respectively. The overall accuracy of change detection was 97% with an absolute error of 0.4 years. We found that the total area of tidal wetlands experienced a net loss of 59.75 km2 in the YRE, but the gain and loss of the study period were 1556.07 and 1615.82 km2, respectively. Land reclamation, sediment reduction, and Spartina alterniflora invasion pose significant threats to tidal wetlands. Sustainable management could be implemented through the establishment of nature reserves and ecological sediment enhancement engineering projects.