Implementation of BFASTmonitor Algorithm on Google Earth Engine to Support Large-Area and Sub-Annual Change Monitoring Using Earth Observation Data

Forest Disturbance Monitoring Using Cloud-Based Sentinel-2 Satellite Imagery and Machine Learning

Forest damage has become more frequent in Hungary in the last decades, and remote sensing offers a powerful tool for monitoring them rapidly and cost-effectively. A combined approach was developed to utilise high-resolution ESA Sentinel-2 satellite imagery and Google Earth Engine cloud computing and field-based forest inventory data. Maps and charts were derived from vegetation indices (NDVI and Z∙NDVI) of satellite images to detect forest disturbances in the Hungarian study site for the period of 2017-2020. The NDVI maps were classified to reveal forest disturbances, and the cloud-based method successfully showed drought and frost damage in the oak-dominated Nagyerdő forest of Debrecen. Differences in the reactions to damage between tree species were visible on the index maps; therefore, a random forest machine learning classifier was applied to show the spatial distribution of dominant species. An accuracy assessment was accomplished with confusion matrices that compared classified index maps to field-surveyed data, demonstrating 99.1% producer, 71% user, and 71% total accuracies for forest damage and 81.9% for tree species. Based on the results of this study and the resilience of Google Earth Engine, the presented method has the potential to be extended to monitor all of Hungary in a faster, more accurate way using systematically collected field-data, the latest satellite imagery, and artificial intelligence.

Using POI and time series Landsat data to identify and rebuilt surface mining, vegetation disturbance and land reclamation process based on Google Earth Engine

The development of coal resources is necessary, but it has a huge negative impact on land, ecology, and the environment. With the increasing awareness of environmental protection and the requirements of related regulations, the design and practice of reclamation projects run through the mining life cycle and continue for a long time after the coal production. High-precision monitoring of mining disturbance and reclamation, quantifying the degree and time of vegetation disturbance and restoration, is of great significance to minimize the environmental effect of mining. Remote sensing, widely used as efficient monitoring tool, but there is not enough research on disturbance and reclamation monitoring taking into account large-scale areas and high temporal and spatial accuracy. Especially when mining sites remain unknown, how to distinguish the disturbance of coal mining and other human activities affecting the surface land cover has become a challenge. Therefore, this paper proposed a method to reconstruct the time series of mining disturbance and reclamation in a large area by using the POI (point of interest) and Landsat time series images using multiple buffer analysis methods. The process includes: (1) Retrieval of POI in the study area based on the public mining list using Python crawler, and buffering 100 km for preliminary extraction of potential mining areas; (2) Using spectral index mask and random forest algorithm to accurately extract the exposed coal on the Google Earth Engine (GEE) platform; (3) Buffering 10 km to identify the occurrence of disturbance and reclamation, using pixel-based temporal trajectory identification of LandTrendr algorithm under GEE. The method successful detect the change points of surface coal mining disturbance and reclamation in eastern Inner Mongolia of China. The results show that: (1) The method can effectively identify the extent of surface coal mining disturbance and reclamation, and the overall extraction accuracy is 81%. (2) Surface coal mining disturbance in eastern Inner Mongolia was concentrated in 2006-2011. By 2020, the total disturbed area is 627.8 km², with an average annual disturbance of 18.5 km², and the annual maximum disturbance to the ground reached 64.6 km² in 2008. With the total reclaimed area being 236.3 km², the reclamation rate is about 37.6%. This study provides a systematic solution and process for monitoring the disturbance and reclamation of surface coal mining in a large range with little known about the mines' location. It can effectively identify the mining disturbance and reclamation process which can also be extended to other areas, providing a quantitative assessment of mining disturbance and reclamation, which can support further ecological restoration decision-making.

Spatiotemporal Dynamics of Wetland in Dongting Lake Based on Multi-Source Satellite Observation Data during Last Two Decades

Monitoring the dynamics of wetland resources has practical value for wetland protection, restoration and sustainable utilization. Dongting Lake wetland reserves are well known for both their intra-annual and inter-annual dynamic changes due to the effects of natural or human factors. However, most wetland monitoring research has failed to consider the seasonal wetlands, which is the most fragile wetland type, requiring more attention. In this study, we used multi-source time series remote sensing data to monitor three Dongting Lake wetland reserves between 2000 and 2020, and the seasonal wetlands were separated from permanent wetlands. Multispectral and indices time series were generated at 30 m resolution using a two-month composition strategy; the optimal features were then selected using the extension of the Jeffries-Matusita distance (J_Bh) and random forest (RF) importance score; yearly wetland maps were identified using the optimal features and the RF classifier. Results showed that (1) the yearly wetland maps had good accuracy, and the overall accuracy and kappa coefficients of all wetland maps from 2000 to 2020 were above 89.6% and 0.86, respectively. Optimal features selected by J_Bh can improve both computational efficiency and classification accuracy. (2) The acreage of seasonal wetlands varies greatly among multiple years due to inter-annual differences in precipitation and evaporation. (3) Although the total wetland area of the three Dongting Lake wetland reserves remained relatively stable between 2000 and 2020, the acreage of the natural wetland types still decreased by 197.0 km², and the change from natural wetland to human-made wetland (paddy field) contributed the most to this decrease. From the perspective of the ecological community, the human-made wetland has lower ecological function value than natural wetlands, so the balance between economic development and ecological protection in the three Dongting Lake wetland reserves requires further evaluation. The outcomes of this study could improve the understanding of the trends and driving mechanisms of wetland dynamics, which has important scientific significance and application value for the protection and restoration of Dongting Lake wetland reserves.

Seeing the System from Above: The Use and Potential of Remote Sensing for Studying Ecosystem Dynamics

Remote sensing techniques are increasingly used for studying ecosystem dynamics, delivering spatially explicit information on the properties of Earth over large spatial and multi-decadal temporal extents. Yet, there is still a gap between the more technology-driven development of novel remote sensing techniques and their applications for studying ecosystem dynamics. Here, I review the existing literature to explore how addressing these gaps might enable recent methods to overcome longstanding challenges in ecological research. First, I trace the emergence of remote sensing as a major tool for understanding ecosystem dynamics. Second, I examine recent developments in the field of remote sensing that are of particular importance for studying ecosystem dynamics. Third, I consider opportunities and challenges for emerging open data and software policies and suggest that remote sensing is at its most powerful when it is theoretically motivated and rigorously ground-truthed. I close with an outlook on four exciting new research frontiers that will define remote sensing ecology in the upcoming decade.

A Pixel-Based Vegetation Greenness Trend Analysis over the Russian Tundra with All Available Landsat Data from 1984 to 2018

As Arctic warming continues, its impact on vegetation greenness is complex, variable and inherently scale-dependent. Studies with multiple spatial resolution satellite observations, with 30 m resolution included, on tundra greenness have been implemented all over the North American tundra. However, finer resolution studies on the greenness trends in the Russian tundra have only been carried out at a limited local or regional scale and the spatial heterogeneity of the trend remains unclear. Here, we analyzed the fine spatial resolution dataset Landsat archive from 1984 to 2018 over the entire Russian tundra and produced pixel-by-pixel greenness trend maps with the support of Google Earth Engine (GEE). The entire Russian tundra was divided into six geographical regions based on World Wildlife Fund (WWF) ecoregions. A Theil–Sen regression (TSR) was used for the trend identification and the changed pixels with a significance level p < 0.05 were retained in the final results for a subsequent greening/browning trend analysis. Our results indicated that: (1) the number of valid Landsat observations was spatially varied. The Western and Eastern European Tundras (WET and EET) had denser observations than other regions, which enabled a trend analysis during the whole study period from 1984 to 2018; (2) the most significant greening occurred in the Yamal-Gydan tundra (WET), Bering tundra and Chukchi Peninsula tundra (CT) during 1984–2018. The EET had a greening trend of 2.3% and 6.6% and the WET of 3.4% and 18% during 1984–1999 and 2000–2018, respectively. The area of browning trend was relatively low when we first masked the surface water bodies out before the trend analysis; and (3) the Landsat-based greenness trend was broadly similar to the AVHRR-based trend over the entire region but AVHRR retrieved more browning areas due to spectral mixing adjacent effects. Higher resolution images and field measurement studies are strongly needed to understand the vegetation trend over the Russian tundra ecosystem.

Identification of Silvicultural Practices in Mediterranean Forests Integrating Landsat Time Series and a Single Coverage of ALS Data

Understanding forest dynamics at the stand level is crucial for sustainable management. Landsat time series have been shown to be effective for identification of drastic changes, such as natural disturbances or clear-cuts, but detecting subtle changes requires further research. Time series of six Landsat-derived vegetation indexes (VIs) were analyzed with the BFAST (Breaks for Additive Season and Trend) algorithm aiming to characterize the changes resulting from harvesting practices of different intensities (clear-cutting, cutting with seed-trees, and thinning) in a Mediterranean forest area of Spain. To assess the contribution of airborne laser scanner (ALS) data and the potential implications of it being after or before the detected changes, two scenarios were defined (based on the year in which ALS data were acquired (2010), and thereby detecting changes from 2005 to 2010 (before ALS data) and from 2011 to 2016 (after ALS data). Pixels identified as change by BFAST were attributed with change in VI intensity and ALS-derived statistics (99th height percentile and forest canopy cover) for classification with random forests, and derivation of change maps. Fusion techniques were applied to leverage the potential of each individual VI change map and to reduce mapping errors. The Tasseled Cap Brightness (TCB) and Normalized Burn Ratio (NBR) indexes provided the most accurate results, the latter being more precise for thinning detection. Our results demonstrate the suitability of Landsat time series and ALS data to characterize forest stand changes caused by harvesting practices of different intensity, with improved accuracy when ALS data is acquired after the change occurs. Clear-cuttings were more readily detectable compared to cutting with seed-trees and thinning, detection of which required fusion approaches. This methodology could be implemented to produce annual cartography of harvesting practices, enabling more accurate statistics and spatially explicit identification of forest operations.

BFAST Lite: A Lightweight Break Detection Method for Time Series Analysis

BFAST Lite is a newly proposed unsupervised time series change detection algorithm that is derived from the original BFAST (Breaks for Additive Season and Trend) algorithm, focusing on improvements to speed and flexibility. The goal of the BFAST Lite algorithm is to aid the upscaling of BFAST for global land cover change detection. In this paper, we introduce and describe the algorithm and then compare its accuracy, speed and features with other algorithms in the BFAST family: BFAST and BFAST Monitor. We tested the three algorithms on an eleven-year-long time series of MODIS imagery, using a global reference dataset with over 30,000 point locations of land cover change to validate the results. We set the parameters of all algorithms to comparable values and analysed the algorithm accuracy over a range of time series ordered by the certainty of that the input time series has at least one abrupt break. To compare the algorithm accuracy, we analysed the time difference between the detected breaks and the reference data to obtain a confusion matrix and derived statistics from it. Lastly, we compared the processing speed of the algorithms using both the original R code as well as an optimised C++ implementation for each algorithm. The results showed that BFAST Lite has similar accuracy to BFAST but is significantly faster, more flexible and can handle missing values. Its ability to use alternative information criteria to select the number of breaks resulted in the best balance between the user’s and producer’s accuracy of detected changes of all the tested algorithms. Therefore, BFAST Lite is a useful addition to the BFAST family of unsupervised time series break detection algorithms, which can be used as an aid in narrowing down areas with changes for updating land cover maps, detecting disturbances or estimating magnitudes and rates of change over large areas.

A Survey on Change Detection and Time Series Analysis with Applications

With the advent of the digital computer, time series analysis has gained wide attention and is being applied to many fields of science. This paper reviews many traditional and recent techniques for time series analysis and change detection, including spectral and wavelet analyses with their advantages and weaknesses. First, Fourier and least-squares-based spectral analysis methods and spectral leakage attenuation methods are reviewed. Second, several time-frequency decomposition methods are described in detail. Third, several change or breakpoints detection methods are briefly reviewed. Finally, some of the applications of the methods in various fields, such as geodesy, geophysics, remote sensing, astronomy, hydrology, finance, and medicine, are listed in a table. The main focus of this paper is reviewing the most recent methods for analyzing non-stationary time series that may not be sampled at equally spaced time intervals without the need for any interpolation prior to the analysis. Understanding the methods presented herein is worthwhile to further develop and apply them for unraveling our universe.

How BFAST Trend and Seasonal Model Components Affect Disturbance Detection in Tropical Dry Forest and Temperate Forest

Time series analysis has gained popularity in forest disturbance monitoring thanks to the availability of satellite and airborne remote sensing images and the development of different time series methods for change detection. Previous research has focused on time series data noise reduction, the magnitude of breakpoints, and accuracy assessment; however, few have looked in detail at how the trend and seasonal model components contribute to disturbance detection in different forest types. Here, we use Landsat time series images spanning 1994–2018 to map forest disturbance in a western Pacific area of Mexico, where both temperate and tropical dry forests have been subject to severe deforestation and forest degradation processes. Since these two forest types have distinct seasonal characteristics, we investigate how trend and seasonal model components, such as the goodness-of-fit (R2), magnitude of change, amplitude, and model length in a stable historical period, affect forest disturbance detection. We applied the Breaks For Additive Season and Trend Monitor (BFAST) algorithm and after accuracy assessment by stratified random sample points, and we obtained 68% and 86% of user accuracy and 75.6% and 86% of producer’s accuracy in disturbance detection, in tropical dry forests and temperate forests, respectively. We extracted the noncorrelated trend and seasonal model components R2, magnitude, amplitude, length of the stable historical period, and percentage of pixels with NA and tested their effects on disturbance detection employing forest-type specific logistic regression. Our results showed that, for all forests combined, the amplitude and stable historical period length contributed to disturbance detection. While for tropical dry forest alone, amplitude was the main predictor, and for the temperate forest alone, the stable historical period length contributed most to the prediction, although it was not statistically significant. These findings provide insights for improving the results of forest disturbance detection in different forest types.

Detecting Vegetation Change in the Pearl River Delta Region Based on Time Series Segmentation and Residual Trend Analysis (TSS-RESTREND) and MODIS NDVI

Time Series Segmentation and Residual Trend analysis (TSS-RESTREND) can detect an abrupt change that was undetected by Residual Trend analysis (RESTREND), but it is usually combined with the Global Inventory for Mapping and Modeling Studies (GIMMS) Normalized Difference Vegetation Index (NDVI), which cannot detect detailed vegetation changes in small areas. Hence, we used Time Series Segmentation and Residual Trend analysis (TSS-RESTREND) and Moderate Resolution Imaging Spectroradiometer (MODIS) NDVI (MOD-TR) to analyze the vegetation dynamic of the Pearl River Delta region (PRD) in this study. To choose the most suitable MODIS NDVI from MOD13Q1 (250 m), MOD13A1 (500 m), and MOD13A2 (1 km), whole and local comparison of results of the break year and MOD-TR were used. Meanwhile, a comparison of vegetation change at the city-scale was also implemented. Moreover, to reduce insignificant trend pixels in TSS-RESTREND, a combination method of TSS-RESTREND and RESTREND (CTSS-RESTREND) was proposed. We found that: (1) MOD13Q1 and MOD13A1 two NDVI were suitable for combination with TSS-RESTREND to detect vegetation change in PRD, but MOD13Q1 was a better choice when considering the accuracy of local detailed vegetation change; (2) CTSS-RESTREND could detect more pixels with a significant change (i.e., significant increase and significant decrease) than those of TSS-RESTREND and RESTREND. Also, its effectiveness could be verified by Landsat data; (3) at the city-scale, the CTSS-RESTREND detected that only vegetation decreases in Shenzhen, Foshan, Dongguan, and Zhongshan were higher than vegetation increases, but, significant vegetation changes (i.e., decreases and increases) were mainly concentrated in Huizhou, Jiangmen, Zhaoqing, and Guangzhou.

Change Detection within Remotely Sensed Satellite Image Time Series via Spectral Analysis

Jump or break detection within a non-stationary time series is a crucial and challenging problem in a broad range of applications including environmental monitoring. Remotely sensed time series are not only non-stationary and unequally spaced (irregularly sampled) but also noisy due to atmospheric effects, such as clouds, haze, and smoke. To address this challenge, a robust method of jump detection is proposed based on the Anti-Leakage Least-Squares Spectral Analysis (ALLSSA) along with an appropriate temporal segmentation. This method, namely, Jumps Upon Spectrum and Trend (JUST), can simultaneously search for trends and statistically significant spectral components of each time series segment to identify the potential jumps by considering appropriate weights associated with the time series. JUST is successfully applied to simulated vegetation time series with varying jump location and magnitude, the number of observations, seasonal component, and noises. Using a collection of simulated and real-world vegetation time series in southeastern Australia, it is shown that JUST performs better than Breaks For Additive Seasonal and Trend (BFAST) in identifying jumps within the trend component of time series with various types. Furthermore, JUST is applied to Landsat 8 composites for a forested region in California, U.S., to show its potential in characterizing spatial and temporal changes in a forested landscape. Therefore, JUST is recommended as a robust and alternative change detection method which can consider the observational uncertainties and does not require any interpolations and/or gap fillings.