Combined Landsat and L-Band SAR Data Improves Land Cover Classification and Change Detection in Dynamic Tropical Landscapes

Monitoring tree canopy dynamics across heterogeneous urban habitats: A longitudinal study using multi-source remote sensing data

Urban trees have attracted increasing attention to serve as a green prescription for addressing various challenges facing human society like climate change and environmental deterioration. However, without healthy growth of urban trees, they cannot service any environmental, social, and economic benefits in a sustainable manner. By monitoring the canopy development, the tree growth dynamics in different urban habitats can be detected and appropriate management approaches can be executed. Using the Kowloon Peninsula, Hong Kong, as a case, this study explores how remote sensing data can help monitor and understand the impacts of heterogeneous urban habitats on tree canopy dynamics. Four algorithms based on WorldView-2 satellite image are compared to optimize the canopy segmentation. Then the individual tree canopy is integrated with Sentinel-2 satellite data to obtain canopy growth dynamics for each season from 2016 to 2020. Three indicators are applied to reflect tree canopy status, including the fluorescence correction vegetation index (FCVI, tracking leaf chlorophyll density), the soil adjusted total vegetation index (SATVI, measuring the density of woody branches and twigs), and the normalised difference phenology index (NDPI, capturing canopy water content). And four heterogeneous habitats where urban trees stand are specified. The results revealed that urban trees show varying canopy growth status, in a descending order from natural terrains, parks, residential lands, to road verges, suggesting that urban habitats curtail trees' growth significantly. Additionally, two super-typhoons in 2017 and 2018, respectively, caused serious damages to tree canopy. Relevant resiliency of tree varies, echoing the sequence of canopy growth status with those in road verges the least resilient. This study shows how remote sensing data can be used to provide a better understanding of long-term tree canopy dynamics across large-scale heterogeneous urban habitats, which is key to monitoring and maintaining the health and growth of urban trees.

Mapping the Spatial Distribution of Fern Thickets and Vine-Laden Forests in the Landscape of Bornean Logged-Over Tropical Secondary Rainforests

Forest degradation has been most frequently defined as an anthropogenic reduction in biomass compared with reference biomass in extant forests. However, so-defined “degraded forests” may widely vary in terms of recoverability. A prolonged loss of recoverability, commonly described as a loss of resilience, poses a true threat to global environments. In Bornean logged-over forests, dense thickets of ferns and vines have been observed to cause arrested secondary succession, and their area may indicate the extent of slow biomass recovery. Therefore, we aimed to discriminate the fern thickets and vine-laden forests from those logged-over forests without dense ferns and vines, as well as mapping their distributions, with the aid of Landsat-8 satellite imagery and machine learning modeling. During the process, we tested whether the gray-level co-occurrence matrix (GLCM) textures of Landsat data and Sentinel-1 C-band SAR data were helpful for this classification. Our study sites were Deramakot and Tangkulap Forest Reserves—commercial production forests in Sabah, Malaysian Borneo. First, we flew drones and obtained aerial images that were used as ground truth for the supervised classification. Subsequently, a machine-learning model with a gradient-boosting decision tree was iteratively tested in order to derive the best model for the classification of the vegetation. Finally, the best model was extrapolated to the entire forest reserve and used to map three classes of vegetation (fern thickets, vine-laden forests, and logged-over forests without ferns and vines) and two non-vegetation classes (bare soil and open water). The overall classification accuracy of the best model was 86.6%; however, by combining the fern and vine classes into the same category, the accuracy was improved to 91.5%. The GLCM texture variables were especially effective at separating fern/vine vegetation from the non-degraded forest, but the SAR data showed a limited effect. Our final vegetation map showed that 30.7% of the reserves were occupied by ferns or vines, which may lead to arrested succession. Considering that our study site was once certified as a well-managed forest, the area of degraded forests with a high risk of loss of resilience is expected to be much broader in other Bornean production forests.

Transforming 2D Radar Remote Sensor Information from a UAV into a 3D World-View

Since unmanned aerial vehicles (UAVs) have been established in geoscience as a key and accessible tool, a wide range of applications are currently being developed. However, not only the design of UAVs themselves is vital to carry out an accurate investigation, but also the sensors and the data processing are key parts to be considered. Several publications including accurate sensors are taking part in pioneer research programs, but less is explained about how they were designed. Besides the commonly used sensors such as a camera, one of the most popular ones is radar. The advantages of a radar sensor to perform research in geosciences are the robustness, the ability to consider large distances and velocity measurements. Unfortunately, these sensors are often expensive and there is a lack of methodological papers that explain how to reduce these costs. To fill this gap, this article aims to show how: (i) we used a radar sensor from the automotive field; and (ii) it is possible to reconstruct a three-dimensional scenario with a UAV and a radar sensor. Our methodological approach proposes a total of eleven stages to process the radar data. To verify and validate the process, a real-world scenario reconstruction is presented with a system resolution reaching from two to three times the radar resolution. We conclude that this research will help the scientific community to include the use of radars in their research projects and programs, reducing costs and increasing accuracy.

3MRS: An Effective Coarse-to-Fine Matching Method for Multimodal Remote Sensing Imagery

The fusion of image data from multiple sensors is crucial for many applications. However, there are significant nonlinear intensity deformations between images from different kinds of sensors, leading to matching failure. To address this need, this paper proposes an effective coarse-to-fine matching method for multimodal remote sensing images (3MRS). In the coarse matching stage, feature points are first detected on a maximum moment map calculated with a phase congruency model. Then, feature description is conducted using an index map constructed by finding the index of the maximum value in all orientations of convolved images obtained using a set of log-Gabor filters. At last, several matches are built through image matching and outlier removal, which can be used to estimate a reliable affine transformation model between the images. In the stage of fine matching, we develop a novel template matching method based on the log-Gabor convolution image sequence and match the template features with a 3D phase correlation matching strategy, given that the initial correspondences are achieved with the estimated transformation. Results show that compared with SIFT, and three state-of-the-art methods designed for multimodal image matching, PSO-SIFT, HAPCG, and RIFT, only 3MRS successfully matched all six types of multimodal remote sensing image pairs: optical–optical, optical–infrared, optical–depth, optical–map, optical–SAR, and day–night, with each including ten different image pairs. On average, the number of correct matches (NCM) of 3MRS was 164.47, 123.91, 4.88, and 4.33 times that of SIFT, PSO-SIFT, HAPCG, and RIFT for the successfully matched image pairs of each method. In terms of accuracy, the root-mean-square error of correct matches for 3MRS, SIFT, PSO-SIFT, HAPCG, and RIFT are 1.47, 1.98, 1.79, 2.83, and 2.45 pixels, respectively, revealing that 3MRS got the highest accuracy. Even though the total running time of 3MRS was the longest, the efficiency for obtaining one correct match is the highest considering the most significant number of matches. The source code of 3MRS and the experimental datasets and detailed results are publicly available.

Mapping Winter Wheat with Optical and SAR Images Based on Google Earth Engine in Henan Province, China

The timely and accurate acquisition of winter wheat acreage is crucial for food security. This study investigated the feasibility of extracting the spatial distribution map of winter wheat in Henan Province by using synthetic aperture radar (SAR, Sentinel-1A) and optical (Sentinel-2) images. Firstly, the SAR images were aggregated based on the growth period of winter wheat, and the optical images were aggregated based on the moderate resolution imaging spectroradiometer normalized difference vegetation index (MODIS-NDVI) curve. Then, five spectral features, two polarization features, and four texture features were selected as feature variables. Finally, the Google Earth Engine (GEE) cloud platform was employed to extract winter wheat acreage through the random forest (RF) algorithm. The results show that: (1) aggregated images based on the growth period of winter wheat and sensor characteristics can improve the mapping accuracy and efficiency; (2) the extraction accuracy of using only SAR images was improved with the accumulation of growth period. The extraction accuracy of using the SAR images in the full growth period reached 80.1%; and (3) the identification effect of integrated images was relatively good, which makes up for the shortcomings of SAR and optical images and improves the extraction accuracy of winter wheat.

Detecting cocoa plantations in Côte d'Ivoire and Ghana and their implications on protected areas

Côte d'Ivoire and Ghana are the largest producers of cocoa in the world. In recent decades the cultivation of this crop has led to the loss of vast tracts of forest areas in both countries. Efficient and accurate methods for remotely identifying cocoa plantations are essential to the implementation of sustainable cocoa practices and for the periodic and effective monitoring of forests. In this study, a method for cocoa plantation identification was developed based on a multi-temporal stack of Sentinel-1 and Sentinel-2 images and a multi-feature Random Forest (RF) algorithm. The Normalized Difference Vegetation Index (NDVI) and second-order texture features were assessed for their importance in an RF classification, and their optimal combination was used as input variables for the RF model to identify cocoa plantations in both countries. The RF model-based cocoa map achieved 82.89% producer's and 62.22% user's accuracy, detecting 3.69 million hectares (Mha) and 2.15 Mha of cocoa plantations for Côte d'Ivoire and Ghana, respectively. The results demonstrate that a combination of an RF model and multi-feature classification can distinguish cocoa plantations from other land cover/use, effectively reducing feature dimensions and improving classification efficiency. The results also highlight that cocoa farms largely encroach into protected areas (PAs), as 20% of the detected cocoa plantation area is located in PAs and almost 70% of the PAs in the study area house cocoa plantations.

Exploiting High Geopositioning Accuracy of SAR Data to Obtain Accurate Geometric Orientation of Optical Satellite Images

Accurate geopositioning of optical satellite imagery is a fundamental step for many photogrammetric applications. Considering the imaging principle and data processing manner, SAR satellites can achieve high geopositioning accuracy. Therefore, SAR data can be a reliable source for providing control information in the orientation of optical satellite images. This paper proposes a practical solution for an accurate orientation of optical satellite images using SAR reference images to take advantage of the merits of SAR data. Firstly, we propose an accurate and robust multimodal image matching method to match the SAR and optical satellite images. This approach includes the development of a new structural-based multimodal applicable feature descriptor that employs angle-weighted oriented gradients (AWOGs) and the utilization of a three-dimensional phase correlation similarity measure. Secondly, we put forward a general optical satellite imagery orientation framework based on multiple SAR reference images, which uses the matches of the SAR and optical satellite images as virtual control points. A large number of experiments not only demonstrate the superiority of the proposed matching method compared to the state-of-the-art methods but also prove the effectiveness of the proposed orientation framework. In particular, the matching performance is improved by about 17% compared with the latest multimodal image matching method, namely, CFOG, and the geopositioning accuracy of optical satellite images is improved, from more than 200 to around 8 m.

Testing the Contribution of Multi-Source Remote Sensing Features for Random Forest Classification of the Greater Amanzule Tropical Peatland

Tropical peatlands such as Ghana's Greater Amanzule peatland are highly valuable ecosystems and under great pressure from anthropogenic land use activities. Accurate measurement of their occurrence and extent is required to facilitate sustainable management. A key challenge, however, is the high cloud cover in the tropics that limits optical remote sensing data acquisition. In this work we combine optical imagery with radar and elevation data to optimise land cover classification for the Greater Amanzule tropical peatland. Sentinel-2, Sentinel-1 and Shuttle Radar Topography Mission (SRTM) imagery were acquired and integrated to drive a machine learning land cover classification using a random forest classifier. Recursive feature elimination was used to optimize high-dimensional and correlated feature space and determine the optimal features for the classification. Six datasets were compared, comprising different combinations of optical, radar and elevation features. Results showed that the best overall accuracy (OA) was found for the integrated Sentinel-2, Sentinel-1 and SRTM dataset (S2+S1+DEM), significantly outperforming all the other classifications with an OA of 94%. Assessment of the sensitivity of land cover classes to image features indicated that elevation and the original Sentinel-1 bands contributed the most to separating tropical peatlands from other land cover types. The integration of more features and the removal of redundant features systematically increased classification accuracy. We estimate Ghana's Greater Amanzule peatland covers 60,187 ha. Our proposed methodological framework contributes a robust workflow for accurate and detailed landscape-scale monitoring of tropical peatlands, while our findings provide timely information critical for the sustainable management of the Greater Amanzule peatland.

Qualifying Land Use and Land Cover Dynamics and Their Impacts on Ecosystem Service in Central Himalaya Transboundary Landscape Based on Google Earth Engine

Land use and land cover (LULC) changes are regarded as one of the key drivers of ecosystem services degradation, especially in mountain regions where they may provide various ecosystem services to local livelihoods and surrounding areas. Additionally, ecosystems and habitats extend across political boundaries, causing more difficulties for ecosystem conservation. LULC in the Kailash Sacred Landscape (KSL) has undergone obvious changes over the past four decades; however, the spatiotemporal changes of the LULC across the whole of the KSL are still unclear, as well as the effects of LULC changes on ecosystem service values (ESVs). Thus, in this study we analyzed LULC changes across the whole of the KSL between 2000 and 2015 using Google Earth Engine (GEE) and quantified their impacts on ESVs. The greatest loss in LULC was found in forest cover, which decreased from 5443.20 km2 in 2000 to 5003.37 km2 in 2015 and which mainly occurred in KSL-Nepal. Meanwhile, the largest growth was observed in grassland (increased by 548.46 km2), followed by cropland (increased by 346.90 km2), both of which mainly occurred in KSL-Nepal. Further analysis showed that the expansions of cropland were the major drivers of the forest cover change in the KSL. Furthermore, the conversion of cropland to shrub land indicated that farmland abandonment existed in the KSL during the study period. The observed forest degradation directly influenced the ESV changes in the KSL. The total ESVs in the KSL decreased from 36.53 × 108 USD y−1 in 2000 to 35.35 × 108 USD y−1 in 2015. Meanwhile, the ESVs of the forestry areas decreased by 1.34 × 108 USD y−1. This shows that the decrease of ESVs in forestry was the primary cause to the loss of total ESVs and also of the high elasticity. Our findings show that even small changes to the LULC, especially in forestry areas, are noteworthy as they could induce a strong ESV response.

Synergistic Use of Sentinel-1 and Sentinel-2 to Map Natural Forest and Acacia Plantation and Stand Ages in North-Central Vietnam

Many remote sensing studies do not distinguish between natural and planted forests. We combine C-Band Synthetic Aperture Radar (Sentinel-1, S-1) and optical satellite imagery (Sentinel-2, S-2) and examine Random Forest (RF) classification of acacia plantations and natural forest in North-Central Vietnam. We demonstrate an ability to distinguish plantation from natural forest, with overall classification accuracies of 87% for S-1, and 92.5% and 92.3% for S-2 and for S-1 and S-2 combined respectively. We found that the ratio of the Short-Wave Infrared Band to the Red Band proved most effective in distinguishing acacia from natural forest. We used RF on S-2 imagery to classify acacia plantations into 6 age classes with an overall accuracy of 70%, with young plantation consistently separated from older. However, accuracy was lower at distinguishing between the older age classes. For both distinguishing plantation and natural forest, and determining plantation age, a combination of radar and optical imagery did nothing to improve classification accuracy.

Drivers of deforestation and degradation for 28 tropical conservation landscapes

Analysing the drivers of deforestation and forest degradation in conservation landscapes can provide crucial information for conservation management. While rates of forest loss can be measured through remote sensing, on the ground information is needed to confirm the commodities and actors behind deforestation. We administered a questionnaire to Wildlife Conservation Society's landscape managers to assess the deforestation drivers in 28 tropical conservation landscapes. Commercial and subsistence agriculture were the main drivers of deforestation, followed by settlement expansion and infrastructure development. Rice, rubber, cassava and maize were the crops most frequently cited as drivers of deforestation in these emblematic conservation landscapes. Landscape managers expected deforestation trends to continue at similar or greater magnitude in the future, calling for urgent measures to mitigate these trends.

Discriminating Forest Successional Stages, Forest Degradation, and Land Use in Central Amazon Using ALOS/PALSAR-2 Full-Polarimetric Data

We discriminated different successional forest stages, forest degradation, and land use classes in the Tapajós National Forest (TNF), located in the Central Brazilian Amazon. We used full polarimetric images from ALOS/PALSAR-2 that have not yet been tested for land use and land cover (LULC) classification, neither for forest degradation classification in the TNF. Our specific objectives were: (1) to test the potential of ALOS/PALSAR-2 full polarimetric images to discriminate LULC classes and forest degradation; (2) to determine the optimum subset of attributes to be used in LULC classification and forest degradation studies; and (3) to evaluate the performance of Random Forest (RF) and Support Vector Machine (SVM) supervised classifications to discriminate LULC classes and forest degradation. PALSAR-2 images from 2015 and 2016 were processed to generate Radar Vegetation Index, Canopy Structure Index, Volume Scattering Index, Biomass Index, and Cloude–Pottier, van Zyl, Freeman–Durden, and Yamaguchi polarimetric decompositions. To determine the optimum subset, we used principal component analysis in order to select the best attributes to discriminate the LULC classes and forest degradation, which were classified by RF. Based on the variable importance score, we selected the four first attributes for 2015, alpha, anisotropy, volumetric scattering, and double-bounce, and for 2016, entropy, anisotropy, surface scattering, and biomass index, subsequently classified by SVM. Individual backscattering indexes and polarimetric decompositions were also considered in both RF and SVM classifiers. Yamaguchi decomposition performed by RF presented the best results, with an overall accuracy (OA) of 76.9% and 83.3%, and Kappa index of 0.70 and 0.80 for 2015 and 2016, respectively. The optimum subset classified by RF showed an OA of 75.4% and 79.9%, and Kappa index of 0.68 and 0.76 for 2015 and 2016, respectively. RF exhibited superior performance in relation to SVM in both years. Polarimetric attributes exhibited an adequate capability to discriminate forest degradation and classes of different ecological succession from the ones with less vegetation cover.

New JAXA High-Resolution Land Use/Land Cover Map for Vietnam Aiming for Natural Forest and Plantation Forest Monitoring

Highly detailed and accurate forest maps are important for various applications including forest monitoring, forestry policy, climate change, and biodiversity loss. This study demonstrates a comprehensive and geographically transferable approach to produce a 12 category high-resolution land use/land cover (LULC) map over mainland Vietnam in 2016 by remote sensing data. The map included several natural forest categories (evergreen broadleaf, deciduous (mostly deciduous broadleaf), and coniferous (mostly evergreen coniferous)) and one category representing all popular plantation forests in Vietnam such as acacia (Acacia mangium, Acacia auriculiformis, Acacia hybrid), eucalyptus (Eucalyptus globulus), rubber (Hevea brasiliensis), and others. The approach combined the advantages of various sensor data by integrating their posterior probabilities resulting from applying a probabilistic classifier (comprised of kernel density estimation and Bayesian inference) to each datum individually. By using different synthetic aperture radar (SAR) images (PALSAR-2/ScanSAR, PALSAR-2 mosaic, Sentinel-1), optical images (Sentinel-2, Landsat-8) and topography data (AW3D30), the resultant map achieved 85.6% for the overall accuracy. The major forest classes including evergreen broadleaf forests and plantation forests had a user’s accuracy and producer’s accuracy ranging from 86.0% to 95.3%. Our map identified 9.55 × 106 ha (±0.16 × 106 ha) of natural forests and 3.89 × 106 ha (±0.11 × 106 ha) of plantation forests over mainland Vietnam, which were close to the Vietnamese government’s statistics (with differences of less than 8%). This study’s result provides a reliable input/reference to support forestry policy and land sciences in Vietnam.

Hkakabo Razi landscape as one of the last exemplar of large contiguous forests

Deforestation and forest degradation around the world endanger the functioning of ecosystems, climate stability, and conservation of biodiversity. We assessed the spatial and temporal dynamics of forest cover in Myanmar’s Hkakabo Razi Landscape (HRL) to determine its integrity based on forest change and fragmentation patterns from 1989 to 2016. Over 80% of the HRL was covered by natural areas, from which forest was the most prevalent (around 60%). Between 1989 and 2016, forest cover declined at an annual rate of 0.225%. Forest degradation occurred mainly around the larger plains of Putao and Naung Mung, areas with relatively high human activity. Although the rate of forest interior loss was approximately 2 to 3 times larger than the rate of total forest loss, forest interior was prevalent with little fragmentation. Physical and environmental variables were the main predictors of either remaining in the current land-cover class or transitioning to another class, although remaining in the current land cover was more likely than land conversion. The forests of the HRL have experienced low human impact and still constitute large tracts of contiguous forest interior. To ensure the protection of these large tracts of forest, sustainable forest policies and management should be implemented.

The Classification Performance and Mechanism of Machine Learning Algorithms in Winter Wheat Mapping Using Sentinel-2 10 m Resolution Imagery

Machine learning algorithms are crucial for crop identification and mapping. However, many works only focus on the identification results of these algorithms, but pay less attention to their classification performance and mechanism. In this paper, based on Google Earth Engine (GEE), Sentinel-2 10 m resolution images during a specific phenological period of winter wheat were obtained. Then, support vector machine (SVM), random forest (RF), and classification and regression tree (CART) machine learning algorithms were employed to identify and map winter wheat in a large-scale area. The hyperparameters of the three machine learning algorithms were tuned by grid search and the 5-fold cross-validation method. The classification performance of the three machine learning algorithms were compared, the results of which demonstrate that SVM achieves best performance in identifying winter wheat, and its overall accuracy (OA), user’s accuracy (UA), producer’s accuracy (PA), and kappa coefficient (Kappa) are 0.94, 0.95, 0.95, and 0.92, respectively. Moreover, 50 various combinations of training and validation sets were used to analyze the generalization ability of the algorithms, and the results show that the average OA of SVM, RF, and CART are 0.93, 0.92, and 0.88, respectively, thus indicating that SVM and RF are more robust than CART. To further explore the sensitivity of SVM, RF, and CART to variations of the algorithm parameters—namely, (C and gamma), (tree and split), and (maxD and minSP)—we employed the grid search method to iterate these parameters, respectively, and to analyze the effect of these parameters on the accuracy scores and classification residuals. It was found that with the change of (C and gamma) in (0.01~1000), SVM’s maximum variation of accuracy score is up to 0.63, and the maximum variation of residuals is 76,215 km2. We concluded that SVM is sensitive to the parameters (C and gamma) and presents a positive correlation. When the parameters (tree and split) change between (100~600) and (1~6), respectively, the RF’s maximum variation of accuracy score is 0.08, and the maximum variation of residuals is 1157 km2, indicating that RF is low in sensitivity toward the parameters (tree and split). When the parameters (maxD and minSP) are between (10~60), the maximum accuracy change value is 0.06, and the maximum variation of residuals is 6943 km2. Therefore, compared to RF, CART is sensitive to the parameters (maxD and minSP) and has poor robustness. In general, under the conditions of the hyperparameters, SVM and RF exhibit optimal classification performance, while CART has relatively inferior performance. Meanwhile, SVM, RF, and CART have different sensitivities toward the algorithm parameters; that is, SVM and CART are more sensitive to the algorithm parameters, while RF has low sensitivity toward changes in the algorithm parameters. The different parameters cause great changes in the accuracy scores and residuals, so it is necessary to determine the algorithm hyperparameters. Generally, default parameters can be used to achieve crop classification, but we recommend the enumeration method, similar to grid search, as a practical way to improve the classification performance of the algorithm if the best classification effect is expected.

Spaceborne L-Band Synthetic Aperture Radar Data for Geoscientific Analyses in Coastal Land Applications: A Review

The coastal zone offers among the world’s most productive and valuable ecosystems and is experiencing increasing pressure from anthropogenic impacts: human settlements, agriculture, aquaculture, trade, industrial activities, oil and gas exploitation and tourism. Earth observation has great capability to deliver valuable data at the local, regional and global scales and can support the assessment and monitoring of land- and water-related applications in coastal zones. Compared to optical satellites, cloud-cover does not limit the timeliness of data acquisition with spaceborne Synthetic Aperture Radar (SAR) sensors, which have all-weather, day and night capabilities. Hence, active radar systems demonstrate great potential for continuous mapping and monitoring of coastal regions, particularly in cloud-prone tropical and sub-tropical climates. The canopy penetration capability with long radar wavelength enables L-band SAR data to be used for coastal terrestrial environments and has been widely applied and investigated for the following geoscientific topics: mapping and monitoring of flooded vegetation and inundated areas; the retrieval of aboveground biomass; and the estimation of soil moisture. Human activities, global population growth, urban sprawl and climate change-induced impacts are leading to increased pressure on coastal ecosystems causing land degradation, deforestation and land use change. This review presents a comprehensive overview of existing research articles that apply spaceborne L-band SAR data for geoscientific analyses that are relevant for coastal land applications.

Post-Disaster Recovery Monitoring with Google Earth Engine

Post-disaster recovery is a complex process in terms of measuring its progress after a disaster and understanding its components and influencing factors. During this process, disaster planners and governments need reliable information to make decisions towards building the affected region back to normal (pre-disaster), or even improved, conditions. Hence, it is essential to use methods to understand the dynamics/variables of the post-disaster recovery process, and rapid and cost-effective data and tools to monitor the process. Google Earth Engine (GEE) provides free access to vast amounts of remote sensing (RS) data and a powerful computing environment in a cloud platform, making it an attractive tool to analyze earth surface data. In this study we assessed the suitability of GEE to analyze and track recovery. To do so, we employed GEE to assess the recovery process over a three-year period after Typhoon Haiyan, which struck Leyte island, in the Philippines, in 2013. We developed an approach to (i) generate cloud and shadow-free image composites from Landsat 7 and 8 satellite imagery and produce land cover classification data using the Random Forest method, and (ii) generate damage and recovery maps based on post-classification change analysis. The method produced land cover maps with accuracies >88%. We used the model to produce damage and three time-step recovery maps for 62 municipalities on Leyte island. The results showed that most of the municipalities had recovered after three years in terms of returning to the pre-disaster situation based on the selected land cover change analysis. However, more analysis (e.g., functional assessment) based on detailed data (e.g., land use maps) is needed to evaluate the more complex and subtle socio-economic aspects of the recovery. The study showed that GEE has good potential for monitoring the recovery process for extensive regions. However, the most important limitation is the lack of very-high-resolution RS data that are critical to assess the process in detail, in particular in complex urban environments.

Land Use/Land Cover Mapping Using Multitemporal Sentinel-2 Imagery and Four Classification Methods—A Case Study from Dak Nong, Vietnam

Information on land use and land cover (LULC) including forest cover is important for the development of strategies for land planning and management. Satellite remotely sensed data of varying resolutions have been an unmatched source of such information that can be used to produce estimates with a greater degree of confidence than traditional inventory estimates. However, use of these data has always been a challenge in tropical regions owing to the complexity of the biophysical environment, clouds, and haze, and atmospheric moisture content, all of which impede accurate LULC classification. We tested a parametric classifier (logistic regression) and three non-parametric machine learning classifiers (improved k-nearest neighbors, random forests, and support vector machine) for classification of multi-temporal Sentinel 2 satellite imagery into LULC categories in Dak Nong province, Vietnam. A total of 446 images, 235 from the year 2017 and 211 from the year 2018, were pre-processed to gain high quality images for mapping LULC in the 6516 km2 study area. The Sentinel 2 images were tested and classified separately for four temporal periods: (i) dry season, (ii) rainy season, (iii) the entirety of the year 2017, and (iv) the combination of dry and rainy seasons. Eleven different LULC classes were discriminated of which five were forest classes. For each combination of temporal image set and classifier, a confusion matrix was constructed using independent reference data and pixel classifications, and the area on the ground of each class was estimated. For overall temporal periods and classifiers, overall accuracy ranged from 63.9% to 80.3%, and the Kappa coefficient ranged from 0.611 to 0.813. Area estimates for individual classes ranged from 70 km2 (1% of the study area) to 2200 km2 (34% of the study area) with greater uncertainties for smaller classes.

Mapping and Assessing the Dynamics of Shifting Agricultural Landscapes Using Google Earth Engine Cloud Computing, a Case Study in Mozambique

Land cover maps obtained at high spatial and temporal resolutions are necessary to support monitoring and management applications in areas with many smallholder and low-input agricultural systems, as those characteristic in Mozambique. Various regional and global land cover products based on Earth Observation data have been developed and made publicly available but their application in regions characterized by a large variety of agro-systems with a dynamic nature is limited by several constraints. Challenges in the classification of spatially heterogeneous landscapes, as in Mozambique, include the definition of the adequate spatial resolution and data input combinations for accurately mapping land cover. Therefore, several combinations of variables were tested for their suitability as input for random forest ensemble classifier aimed at mapping the spatial dynamics of smallholder agricultural landscape in Vilankulo district in Mozambique. The variables comprised spectral bands from Landsat 7 ETM+ and Landsat 8 OLI/TIRS, vegetation indices and textural features and the classification was performed within the Google Earth Engine cloud computing for the years 2012, 2015, and 2018. The study of three different years aimed at evaluating the temporal dynamics of the landscape, typically characterized by high shifting nature. For the three years, the best performing variables included three selected spectral bands and textural features extracted using a window size of 25. The classification overall accuracy was 0.94 for the year 2012, 0.98 for 2015, and 0.89 for 2018, suggesting that the produced maps are reliable. In addition, the areal statistics of the class classified as agriculture were very similar to the ground truth data as reported by the Serviços Distritais de Actividades Económicas (SDAE), with an average percentage deviation below 10%. When comparing the three years studied, the natural vegetation classes are the predominant covers while the agriculture is the most important cause of land cover changes.

Combining Radar and Optical Imagery to Map Oil Palm Plantations in Sumatra, Indonesia, Using the Google Earth Engine

Monitoring the expansion of commodity crops in the tropics is crucial to safeguard forests for biodiversity and ecosystem services. Oil palm (Elaeis guineensis) is one such crop that is a major driver of deforestation in Southeast Asia. We evaluated the use of a semi-automated approach with random forest as a classifier and combined optical and radar datasets to classify oil palm land-cover in 2015 in Sumatra, Indonesia, using Google Earth Engine. We compared our map with two existing remotely-sensed oil palm land-cover products that utilized visual and semi-automated approaches for the same year. We evaluated the accuracy of oil palm land-cover classification from optical (Landsat), radar (synthetic aperture radar (SAR)), and combined optical and radar satellite imagery (Combined). Combining Landsat and SAR data resulted in the highest overall classification accuracy (84%) and highest producer’s and user’s accuracy for oil palm classification (84% and 90%, respectively). The amount of oil palm land-cover in our Combined map was closer to official government statistics than the two existing land-cover products that used visual interpretation techniques. Our analysis of the extents of disagreement in oil palm land-cover indicated that our map had comparable accuracy to one of them and higher accuracy than the other. Our results demonstrate that a combination of optical and radar data outperforms the use of optical-only or radar-only datasets for oil palm classification and that our technique of preprocessing and classifying combined optical and radar data in the Google Earth Engine can be applied to accurately monitor oil-palm land-cover in Southeast Asia.

Synergistic Use of Radar and Optical Satellite Data for Improved Monsoon Cropland Mapping in India

Monsoon crops play a critical role in Indian agriculture, hence, monitoring these crops is vital for supporting economic growth and food security for the country. However, monitoring these crops is challenging due to limited availability of optical satellite data due to cloud cover during crop growth stages, landscape heterogeneity, and small field sizes. In this paper, our objective is to develop a robust methodology for high-resolution (10 m) monsoon cropland mapping appropriate for different agro-ecological regions (AER) in India. We adapted a synergistic approach of combining Sentinel-1 Synthetic Aperture Radar (SAR) data with Normalized Difference Vegetation Index (NDVI) derived from Sentinel-2 optical data using the Google Earth Engine platform. We developed a new technique, Radar Optical cross Masking (ROM), for separating cropland from non-cropland by masking out forest, plantation, and other non-dynamic features. The methodology was tested for five different AERs in India, representing a wide diversity in agriculture, soil, and climatic variations. Our findings indicate that the overall accuracy obtained by using the SAR-only approach is 90%, whereas that of the combined approach is 93%. Our proposed methodology is particularly effective in regions with cropland mixed with tree plantation/mixed forest, typical of smallholder dominated tropical countries. The proposed agriculture mask, ROM, has high potential to support the global agriculture monitoring missions of Geo Global Agriculture Monitoring (GEOGLAM) and Sentinel-2 for Agriculture (S2Agri) project for constructing a dynamic monsoon cropland mask.

Evaluating remote sensing datasets and machine learning algorithms for mapping plantations and successional forests in Phnom Kulen National Park of Cambodia

This study develops a modelling framework by utilizing multi-sensor imagery for classifying different forest and land use types in the Phnom Kulen National Park (PKNP) in Cambodia. Three remote sensing datasets (Landsat optical data, ALOS L-band data and LiDAR derived Canopy Height Model (CHM)) were used in conjunction with three different machine learning (ML) regression techniques (Support Vector Machines (SVM), Random Forests (RF) and Artificial Neural Networks (ANN)). These ML methods were implemented on (a) Landsat spectral data, (b) Landsat spectral band & ALOS backscatter data, and (c) Landsat spectral band, ALOS backscatter data, & LiDAR CHM data. The Landsat-ALOS combination produced more accurate classification results (95% overall accuracy with SVM) compared to Landsat-only bands for all ML models. Inclusion of LiDAR CHM (which is a proxy for vertical canopy heights) improved the overall accuracy to 98%. The research establishes that majority of PKNP is dominated by cashew plantations and the nearly intact forests are concentrated in the more inaccessible parts of the park. The findings demonstrate how different RS datasets can be used in conjunction with different ML models to map forests that had undergone varying levels of degradation and plantations.

Tracking the Land Use/Land Cover Change in an Area with Underground Mining and Reforestation via Continuous Landsat Classification

Understanding the changes in a land use/land cover (LULC) is important for environmental assessment and land management. However, tracking the dynamic of LULC has proved difficult, especially in large-scale underground mining areas with extensive LULC heterogeneity and a history of multiple disturbances. Additional research related to the methods in this field is still needed. In this study, we tracked the LULC change in the Nanjiao mining area, Shanxi Province, China between 1987 and 2017 via random forest classifier and continuous Landsat imagery, where years of underground mining and reforestation projects have occurred. We applied a Savitzky–Golay filter and a normalized difference vegetation index (NDVI)-based approach to detect the temporal and spatial change, respectively. The accuracy assessment shows that the random forest classifier has a good performance in this heterogeneous area, with an accuracy ranging from 81.92% to 86.6%, which is also higher than that via support vector machine (SVM), neural network (NN), and maximum likelihood (ML) algorithm. LULC classification results reveal that cultivated forest in the mining area increased significantly after 2004, while the spatial extent of natural forest, buildings, and farmland decreased significantly after 2007. The areas where vegetation was significantly reduced were mainly because of the transformation from natural forest and shrubs into grasslands and bare lands, respectively, whereas the areas with an obvious increase in NDVI were mainly because of the conversion from grasslands and buildings into cultivated forest, especially when villages were abandoned after mining subsidence. A partial correlation analysis demonstrated that the extent of LULC change was significantly related to coal production and reforestation, which indicated the effects of underground mining and reforestation projects on LULC changes. This study suggests that continuous Landsat classification via random forest classifier could be effective in monitoring the long-term dynamics of LULC changes, and provide crucial information and data for the understanding of the driving forces of LULC change, environmental impact assessment, and ecological protection planning in large-scale mining areas.

Land Cover Changes and Their Driving Mechanisms in Central Asia from 2001 to 2017 Supported by Google Earth Engine

Limited research has been published on land changes and their driving mechanisms in Central Asia, but this area is an important ecologically sensitive area. Supported by Google Earth Engine (GEE), this study used Landsat satellite imagery and selected the random forest algorithm to perform land classification and obtain the annual land cover datasets of Central Asia from 2001 to 2017. Based on the temporal datasets, the distributions and dynamic trends of land cover were summarized, and the key factors driving land changes were analyzed. The results show that (1) the obtained land datasets are reliable and highly accurate, with an overall accuracy of 0.90 ± 0.01. (2) Grassland and bareland are the two most prominent land cover types, with area proportions of 45.0% and 32.9% in 2017, respectively. Over the past 17 years, bareland has displayed an overall reduction, decreasing by 2.6% overall. Natural vegetation (grassland, forest, and shrubland), cultivated land, water bodies and wetlands have displayed increasing trends at different rates. (3) The amount of precipitation and degree of drought are the driving factors that affect natural vegetation. The changes in cultivated land are mainly affected by precipitation and anthropogenic drivers. The effects of increasing urban populations and expanding industrial development are the factors driving the expansion of urban regions. The advantages and uncertainties arising from the land mapping and change detection method and the complexity of the driving mechanisms are also discussed.

Mapping Annual Forest Change Due to Afforestation in Guangdong Province of China Using Active and Passive Remote Sensing Data

Accurate acquisition of spatial distribution of afforestation in a large area is of great significance to contributing to the sustainable utilization of forest resources and the evaluation of the carbon accounting. Annual forest maps (1986–2016) of Guangdong, China were generated using time series Landsat images and PALSAR data. Initially, four PALSAR-based classifiers were used to classify land cover types. Then, the optimal mapping algorithm was determined. Next, an accurate identification of forest and non-forest was carried out by combining Landsat-based phenological variables and PALSAR-based land cover classifications. Finally, the spatio-temporal distribution of forest cover change due to afforestation was created and its forest biomass dynamics changes were detected. The results indicated that the overall accuracy of forest classification of the improved model based on the PALSAR-based stochastic gradient boosting (SGB) classification and the maximum value of normalized difference vegetation index (NDVI; SGB-NDVI) were approximately 75–85% in 2005, 2010, and 2016. Compared with the Japan Aerospace Exploration Agency (JAXA) PALSAR-forest/non-forest, the SGB-NDVI-based forest product showed great improvement, while the SGB-NDVI product was the same or slightly inferior to the Global Land Cover (GLC) and vegetation tracker change (VCT)-based land cover types, respectively. Although this combination of multiple sources contained some errors, the SGB-NDVI model effectively identified the distribution of forest cover changes by afforestation events. By integrating aboveground biomass dynamics (AGB) change with forest cover, the trend in afforestation area closely corresponded with the trend in forest AGB. This technique can provide an essential data baseline for carbon assessment in the planted forests of southern China.

Integrating Analytical Frameworks to Investigate Land-Cover Regime Shifts in Dynamic Landscapes

Regime shifts—rapid long-term transitions between stable states—are well documented in ecology but remain controversial and understudied in land use and land cover change (LUCC). In particular, uncertainty surrounds the prevalence and causes of regime shifts at the landscape level. We studied LUCC dynamics in the Tanintharyi Region (Myanmar), which contains one of the last remaining significant contiguous forest areas in Southeast Asia but was heavily deforested between 1992–2015. By combining remote sensing methods and a literature review of historical processes leading to LUCC, we identified a regime shift from a forest-oriented state to an agricultural-oriented state between 1997–2004. The regime shift was triggered by a confluence of complex political and economic conditions within Myanmar, notably the ceasefires between various ethnic groups and the military government, coupled with its enhanced business relations with Thailand and China. Government policies and foreign direct investment enabling the establishment of large-scale agro-industrial concessions reinforced the new agriculture-oriented regime and prevented reversion to the original forest-dominated regime. Our approach of integrating complementary analytical frameworks to identify and understand land-cover regime shifts can help policymakers to preempt future regime shifts in Tanintharyi, and can be applied to the study of land change in other regions.

On the Synergistic Use of Optical and SAR Time-Series Satellite Data for Small Mammal Disease Host Mapping

(1) Background: Echinococcus multilocularis (Em), a highly pathogenic parasitic tapeworm, is responsible for a significant burden of human disease. In this study, optical and time-series Synthetic Aperture Radar (SAR) data is used synergistically to model key land cover characteristics driving the spatial distributions of two small mammal intermediate host species, Ellobius tancrei and Microtus gregalis, which facilitate Em transmission in a highly endemic area of Kyrgyzstan. (2) Methods: A series of land cover maps are derived from (a) single-date Landsat Operational Land Imager (OLI) imagery, (b) time-series Sentinel-1 SAR data, and (c) Landsat OLI and time-series Sentinel-1 SAR data in combination. Small mammal distributions are analyzed in relation to the surrounding land cover class coverage using random forests, before being applied predictively over broader areas. A comparison of models derived from the three land cover maps are made, assessing their potential for use in cloud-prone areas. (3) Results: Classification accuracies demonstrated the combined OLI-SAR classification to be of highest accuracy, with the single-date OLI and time-series SAR derived classifications of equivalent quality. Random forest analysis identified statistically significant positive relationships between E. tancrei density and agricultural land, and between M. gregalis density and water and bushes. Predictive application of random forest models identified hotspots of high relative density of E. tancrei and M. gregalis across the broader study area. (4) Conclusions: This offers valuable information to improve the targeting of limited-resource disease control activities to disrupt disease transmission in this area. Time-series SAR derived land cover maps are shown to be of equivalent quality to those generated from single-date optical imagery, which enables application of these methods in cloud-affected areas where, previously, this was not possible due to the sparsity of cloud-free optical imagery.

Assessing environmental impacts and change in Myanmar's mangrove ecosystem service value due to deforestation (2000-2014)

Myanmar is one of the mangrove-richest countries in the world, providing valuable ecosystem services to people. However, due to deforestation driven primarily by agricultural expansion, Myanmar's mangrove forest cover has declined dramatically over the past few decades, while what remains is still under pressure. To support management planning, accurate quantification of mangrove forest cover changes on a national scale is needed. In this study, we quantified Myanmar's mangrove forest cover changes between 2000 and 2014 using remotely sensed data, examined the environmental impacts of such changes, and estimated the changes in the economic values of mangrove ecosystem services in the country. Results indicate that Myanmar had a net mangrove loss of 191,122 ha over the study period. Since 2000, Myanmar has been losing mangrove forest cover at an alarming rate of 14,619 ha/year (2.2%/year). The loss was predominant in Rakhine and Ayeyarwady. The observed mangrove forest cover loss has resulted in decreased evapotranspiration, carbon stock, and tree cover percentage. Due to deforestation, Myanmar also suffered a net loss of 2,397 million US$/year in its mangrove ecosystem service value (i.e. 28.7% decrease from 2000), in which maintenance of fisheries nursery populations and habitat and coastal protection were among those services that were greatly affected. We suggest that intensive reforestation and mangrove protection programs be implemented immediately. Agroforestry and community forestry programs are encouraged in areas that are under immense pressure from paddy field expansion, fuelwood extraction, charcoal production, and fish and shrimp farming activities. Potential alternative sustainable solutions should include intensive government-led private forest plantations or community-owned forest plantations to be developed with care by local farmers, nongovernmental organizations, and business owners.

JAXA High-Resolution Land Use/Land Cover Map for Central Vietnam in 2007 and 2017

Robust remote monitoring of land cover changes is essential for a range of studies such as climate modeling, ecosystems, and environmental protection. However, since each satellite data has its own effective features, it is difficult to obtain high accuracy land cover products derived from a single satellite’s data, perhaps because of cloud cover, suboptimal acquisition schedules, and the restriction of data accessibility. In this study, we integrated Landsat 5, 7, and 8, Sentinel-2, Advanced Land Observing Satellite Advanced Visual, and Near Infrared Radiometer type 2 (ALOS/AVNIR-2), ALOS Phased Array L-band Synthetic Aperture Radar (PALSAR) Mosaic, ALOS-2/PALSAR-2 Mosaic, Shuttle Radar Topography Mission (SRTM), and ancillary data, using kernel density estimation to map and analyze land use/cover change (LUCC) over Central Vietnam from 2007 to 2017. The region was classified into nine categories, i.e., water, urban, rice paddy, upland crops, grassland, orchard, forest, mangrove, and bare land by an automatic model which was trained and tested by 98,000 reference data collected from field surveys and visual interpretations. Results were the 2007 and 2017 classified maps with the same spatial resolutions of 10 m and the overall accuracies of 90.5% and 90.6%, respectively. They indicated that Central Vietnam experienced an extensive change in land cover (33 ± 18% of the total area) during the study period. Gross gains in forests (2680 km2) and water bodies (570 km2) were primarily from conversion of orchards, paddy fields, and crops. Total losses in bare land (495 km2) and paddy (485 km2) were largely to due transformation to croplands and urban & other infrastructure lands. In addition, the results demonstrated that using global land cover products for specific applications is impaired because of uncertainties and inconsistencies. These findings are essential for the development of resource management strategy and environmental studies.