The Google Earth Engine (GEE) portal provides enhanced opportunities for undertaking earth observation studies. Established towards the end of 2010, it provides access to satellite and other ...ancillary data, cloud computing, and algorithms for processing large amounts of data with relative ease. However, the uptake and usage of the opportunity remains varied and unclear. This study was undertaken to investigate the usage patterns of the Google Earth Engine platform and whether researchers in developing countries were making use of the opportunity. Analysis of published literature showed that a total of 300 journal papers were published between 2011 and June 2017 that used GEE in their research, spread across 158 journals. The highest number of papers were in the journal Remote Sensing, followed by Remote Sensing of Environment. There were also a number of papers in premium journals such as Nature and Science. The application areas were quite varied, ranging from forest and vegetation studies to medical fields such as malaria. Landsat was the most widely used dataset; it is the biggest component of the GEE data portal, with data from the first to the current Landsat series available for use and download. Examination of data also showed that the usage was dominated by institutions based in developed nations, with study sites mainly in developed nations. There were very few studies originating from institutions based in less developed nations and those that targeted less developed nations, particularly in the African continent.
The Google Earth Engine (GEE) is a cloud computing platform designed to store and process huge data sets (at petabyte-scale) for analysis and ultimate decision making ...
Accurate measurement and mapping of biomass is a critical component of carbon stock quantification, climate change impact assessment, suitability and location of bio-energy processing plants, ...assessing fuel for forest fires, and assessing merchandisable timber....
•First national scale deep learning-based remote sensing mapping of SOC in South Africa.•Performance of Deep neural network was compared to other classical machine learning models.•Results revealed ...that model type has a significant impact on prediction accuracy.•Sentinel-3 performed relatively well as a single data source for national scale SOC mapping.•Sentinel-3′s high spectral resolution compliments its low spatial resolution.
Mapping of soil organic carbon (SOC) at the regional level is critical for climate change policy and the mitigation of its adverse effects. However, reliable SOC estimates particularly over a large extent remains a major challenge due to among others limited sample points, quality of simulation data and the algorithm adopted. Remote sensing (RS) strategies have emerged as a suitable alternative to field and laboratory SOC determination, especially at large spatial extent. The use of Sentinel-3 sensor, the latest of the Sentinel series is minimal and has not been fully developed, despite its impressive attributes that include high spectral-temporal resolution and large coverage. Compared to linear and classical ML models, deep learning (DL) models offer a considerable improvement in data analysis due to their ability to extract more representative features and identify complex spatial patterns associated with big data. Yet, there is paucity in literature on the application of dl-based remote sensing strategies for SOC prediction. Consequently, this study adopted a deep neural network (DNN) to predict SOC at a national scale, using Sentinel-3 image, and compared the results with random forest (RF), support vector machine (SVM) and artificial neural network (ANN) models. The models were trained based on 10-fold cross-validation with 1936 soil samples and 31 predictors. The DNN model generated the best result with a root mean square error (RMSE) of 10.35 t/ha (26 % of the mean), followed by RF (RMSE = 11.2 t/ha), ANN (RMSE = 11.6 t/ha) and SVM (RMSE = 13.6 t/ha). The analytical prowess of the DNN, together with its ability to handle big data by learning patterns through a series of hidden layers (10) to draw conclusions, gives it an edge over other classical ML models. The study concluded that the DNN model with Sentinel-3 data is promising and provides an effective framework for continuous national level SOC modelling.
The major constraint in understanding grass above ground biomass variations using remotely sensed data are the expenses associated with the data, as well as the limited number of techniques that can ...be applied to different management practices with minimal errors. New generation multispectral sensors such as Sentinel 2 Multispectral Imager (MSI) are promising for effective rangeland management due to their unique spectral bands and higher signal to noise ratio. This study resampled hyperspectral data to spectral resolutions of the newly launched Sentinel 2 MSI and the recently launched Landsat 8 OLI for comparison purposes. Using Sparse partial least squares regression, the resampled data was applied in estimating above ground biomass of grasses treated with different fertilizer combinations of ammonium sulfate, ammonium nitrate, phosphorus and lime as well as unfertilized experimental plots. Sentinel 2 MSI derived models satisfactorily performed (R2=0.81, RMSEP=1.07kg/m2, RMSEP_rel=14.97) in estimating grass above ground biomass across different fertilizer treatments relative to Landsat 8 OLI (Landsat 8 OLI: R2=0.76, RMSEP=1.15kg/m2, RMSEP_rel=16.04). In comparison, hyperspectral data derived models exhibited better grass above ground biomass estimation across complex fertilizer combinations (R2=0.92, RMSEP=0.69kg/m2, RMSEP_rel=9.61). Although Sentinel 2 MSI bands and indices better predicted above ground biomass compared with Landsat 8 OLI bands and indices, there were no significant differences (α=0.05) in the errors of prediction between the two new generational sensors across all fertilizer treatments. The findings of this study portrays Sentinel 2 MSI and Landsat 8 OLI as promising remotely sensed datasets for regional scale biomass estimation, particularly in resource scarce areas.
•Landscape metrics indices and the forest fragmentation model captured long term temporal dynamics of vegetation fragmentation.•Local Indicators of Spatial Autocorrelation (LISA) indices captured ...clustered and dispersed patterns of vegetation patches.•Significant increase in vegetation fragmentation during the time period 1994–2017.•Large positive z-scores of LISA was detected in lowly fragmented vegetated patches in the northern part of the city.•Low and negative z-scores of LISA was associated with highly fragmented vegetation patches in the densely built-up areas.
Rapid urbanization threatens urban green spaces and vegetation, demonstrated by a decrease in connectivity and higher levels of fragmentation. Understanding historic spatial and temporal patterns of such fragmentation is important for habitat and biological conservation, ecosystem management and urban planning. Despite their potential value, Local Indicators of Spatial Autocorrelation (LISA) measures have not been sufficiently exploited in monitoring the spatial and temporal variability in clustering and fragmentation of vegetation patterns in urban areas. LISA statistics are an important structural measure that indicates the presence of outliers, zones of similarity (hot spots) and of dissimilarity (cold spots) at proximate locations, hence they could be used to explicitly capture spatial patterns that are clustered, dispersed or random. In this study, we applied landscape metrics, LISA indices to analyse the temporal variability in clustering and fragmentation patterns of vegetation patches in Harare metropolitan city, Zimbabwe using Landsat series data for 1994, 2001 and 2017. Analysis of landscape metrics showed an increase in the fragmentation of vegetation patches between 1994–2017 as shown by the decrease in mean patch size, an increase in number of patches, edge density and shape complexity of vegetation patches. The study further demonstrates the utility of LISA indices in identifying key hot spot and cold spots. Comparatively, the highly vegetated northern parts of the city were characterised by significantly high positive spatial autocorrelation (p < 0.05) of vegetation patches. Conversely, more dispersed vegetation patches were found in the highly and densely urbanized western, eastern and southern parts of the city. This suggest that with increasing vegetation fragmentation, small and isolated vegetation patches do not spatially cluster but are dispersed geographically. The findings of the study underline the potential of LISA measures as a valuable spatially explicit method for the assessment of spatial clustering and fragmentation of urban vegetation patterns.
Wetland vegetation plays a key role in the ecological functions of wetland environments. Remote sensing techniques offer timely, up-to-date, and relatively accurate information for sustainable and ...effective management of wetland vegetation. This article provides an overview on the status of remote sensing applications in discriminating and mapping wetland vegetation, and estimating some of the biochemical and biophysical parameters of wetland vegetation. Research needs for successful applications of remote sensing in wetland vegetation mapping and the major challenges are also discussed. The review focuses on providing fundamental information relating to the spectral characteristics of wetland vegetation, discriminating wetland vegetation using broad- and narrow-bands, as well as estimating water content, biomass, and leaf area index. It can be concluded that the remote sensing of wetland vegetation has some particular challenges that require careful consideration in order to obtain successful results. These include an in-depth understanding of the factors affecting the interaction between electromagnetic radiation and wetland vegetation in a particular environment, selecting appropriate spatial and spectral resolution as well as suitable processing techniques for extracting spectral information of wetland vegetation.
As the risk of climate change increases, robust fire monitoring methods become critical for fire management purposes. National-scale spatiotemporal patterns of the fires and how they relate to ...vegetation and environmental conditions are not well understood in Zimbabwe. This paper presents a spatially explicit method combining satellite data and spatial statistics in detecting spatiotemporal patterns of fires in Zimbabwe. The Emerging Hot Spot Analysis method was utilized to detect statistically significant spatiotemporal patterns of fire occurrence between the years 2002 and 2021. Statistical analysis was done to determine the association between the spatiotemporal patterns and some environmental variables such as topography, land cover, land use, ecoregions and precipitation. The highest number of fires occurred in September, coinciding with Zimbabwe's observed fire season. The number of fires significantly varied among seasons, with the hot and dry season (August to October) recording the highest fire counts. Additionally, although June, July and November are not part of the official fire season in Zimbabwe, the fire counts recorded for these months were relatively high. This new information has therefore shown the need for revision of the fire season in Zimbabwe. The northern regions were characterized by persistent, oscillating, diminishing and historical spatiotemporal fire hotspots. Agroecological regions IIa and IIb and the Southern Miombo bushveld ecoregion were the most fire-prone areas. The research findings also revealed new critical information about the spatiotemporal fire patterns in various terrestrial ecoregions, land cover, land use, precipitation and topography and highlighted potential areas for effective fire management strategies.
•We conducted a brief review of deep learning-based remote sensing of SOC in the last 20 years.•Most studies are concentrated in the northern hemisphere.•Deep learning models produces better accuracy ...than traditional neural networks models.•Studies utilized more hyperspectral data compared to multispectral and radar.•We finally provided insights on future research directions in deep learning-based SOC remote sensing.
Understanding soil organic carbon (SOC) is critical to, among others, atmospherics and terrestrial carbon balance and climate change mitigation. This has necessitated development of novel analytical approaches to determine SOC. The use of neural network (NN) models for analysis of landscape biophysical and biochemical properties based on remotely sensed data has become popular in the last decade. This has been facilitated by the proliferation of “big data” from earth observation systems as well as the advances in machine learning (ML). Specifically, the use of traditional neural networks (TNN) and transition to deep learning (DL) frameworks offers considerable improvement in the performance and accuracy of SOC retrieval from remotely sensed data. This paper seeks to provide a brief review on the use of TNN and DL-based remote sensing strategies in SOC estimation, with focus on the progression, application, potential and limitations. The review concludes by providing major challenges impeding the wide adoption of DL frameworks in remote sensing applications of SOC, as well as the potential future research direction.
Mapping of patterns and spatial distribution of land-use/cover (LULC) has long been based on remotely sensed data. In the recent past, efforts to improve the reliability of LULC maps have seen a ...proliferation of image classification techniques. Despite these efforts, derived LULC maps are still often judged to be of insufficient quality for operational applications, due to disagreement between generated maps and reference data. In this study we sought to pursue two objectives: first, to test the new-generation multispectral RapidEye imagery classification output using machine-learning random forest (RF) and support vector machines (SVM) classifiers in a heterogeneous coastal landscape; and second, to determine the importance of different RapidEye bands on classification output. Accuracy of the derived thematic maps was assessed by computing confusion matrices of the classifiers' cover maps with respective independent validation data sets. An overall classification accuracy of 93.07% with a kappa value of 0.92, and 91.80 with a kappa value of 0.92 was achieved using RF and SVM, respectively. In this study, RF and SVM classifiers performed comparatively similarly as demonstrated by the results of McNemer's test (Z = 1.15). An evaluation of different RapidEye bands using the two classifiers showed that incorporation of the red-edge band has a significant effect on the overall classification accuracy in vegetation cover types. Consequently, pursuit of high classification accuracy using high-spatial resolution imagery on complex landscapes remains paramount.