Landsat-8 OLI Imagery Research Articles

Vegetation monitoring in mining area using canopy reflectance and landsat8 oli image

The aim of this study is to find out the vegetation monitoring in mining areas which is of great significance in grasping the ecological environment status of mining areas. As the first target of spectral reflection, the vegetation canopy directly affects the results of spectral reception. Leaf Angle Distribution (LAD) is a key canopy structural parameter, which directly determines the amount of solar radiation intercepted by the canopy. Taking the relationship between the effect of leaf inclination angle on vegetation canopy reflectance and NDVI as an entry point, the study monitored and analyzed the vegetation in Dexing mining area, China, from 2013 to 2018 by NDVI, based on which the Copper Stress Vegetation Index (CSVI), which is applicable to the narrow wavelength band of remote sensing imagery, was evaluated whether it could be applied to the wide wavelength band. The results show that the NDVI value of the horizontal distribution is significantly higher than that of other leaf inclination distributions, and the NDVI in the direction of the horizontal distribution is 0.82, while the NDVI in the direction of the hi-straight distribution is 0.75; the NDVI of the mining area decreases gradually with the increase of the area of the mining area, and the correlation coefficients of the two groups of CSVI based on the wide and narrow bands are not large, which is not suitable for the vegetation monitoring by using the Landsat8 OLI imagery. Wide band is less applicable for vegetation monitoring. Bangladesh J. Bot. 53(3): 835-847, 2024 (September) Special

Mapping a tropical regolith architecture: A comparative LANDSAT-8 and SENTINEL-2 analysis

Understanding of the lateritic regolith units is important for regional geological and pedological mapping and provides valuable insights into landscape evolution and mineral exploration, especially in tropical environments. To contribute to these issues, this study used Landsat-8 OLI and Sentinel-2 MSI imagery to map the different lateritic regolith units in an area of 133 thousand km2 in the Midwest of Brazil. The approach used the Directed Principal Component Analysis (DPCA) technique and SRTM data as predictor variables for the Random Forest (RF) classification model, with the aim of comparing the efficiency of both sensors following identical classification methods. The results showed a similar overall accuracy of 73% for Landsat-8 and 74% for Sentinel-2, however the Sentinel-2-based maps are the most accurate with a kappa of 0.63 and a tau of 0.68. The significant levels of accuracy highlight the efficiency and feasibility of the multispectral imagery and machine learning approach to lateritic regolith mapping. It also emphasizes the importance of considering the spectral and spatial characteristics of sensors when selecting the most appropriate one for geological, pedological and geomorphological studies in tropical regions.

Identifying Dominant Structural Pattern of Semarang City Using Digital Elevation Model and Landsat 8-OLI Imagery

Semarang City is an area in the northern part of Java Island, administratively the Capital of Central Java Province. Because Semarang City is the provincial capital, Semarang City has a relatively dense population. Geologically, the city of Semarang is an area that consists of various types of lithology and is traversed by various regional geological structures. In this case, understanding the existence and distribution of geological structures in the city of Semarang is essential, considering that geological structures are one of the controllers for natural disasters such as landslides and earthquakes. This study analyzed the existence and distribution of geological structures in the city of Semarang based on the lineaments observed from digital elevation models and satellite imagery. This study aims to identify the dominant structural pattern in Semarang City, determine the relationship between fault fracture density (FFD) with regional geological structure and lithology, and determine the fault zone area in Semarang City based on FFD & lineament analysis. The method used in this analysis is to process DEM data and Landsat 8-OLI imagery, then interpret the lineaments in the form of rosset diagrams and the density in the form of FFD maps. The results of the rosset diagram analysis show that Semarang City has various structural lineament patterns, namely: North–South, Northeast–Southwest, and Northwest–Southeast, with the North–South pattern as the dominant pattern. Based on the results of the lineament density distribution on the FFD map, it is known that the area traversed by the Semarang regional geological structure has a high lineament density value which is interpreted that the area is a weak zone with high structural intensity. From this study, it can also be seen that there is no significant relationship between the type of lithology and the density value on the FFD map. The distribution of lineament density is not affected by the type of lithology, except in the northern and northeastern parts of Semarang city, which consist of alluvium. Based on these results, it can be interpreted that the fault zone area is associated with areas that have high-density values on the FFD map. Distribution of the fault zone area of Semarang City is spreading over the Banyumanik, Gunungpati, and Mijen Districts, which are relatively in the southern and central parts of Semarang City.

Assessing lake shoreline change and prediction for 2030 by physical drivers: A Case Study from Lake Batur, Batur UNESCO Global Geopark, Bali

Over ten years, the water level of Lake Batur has increased. Agriculture area and settlements around Lake Batur are threatened by rising water level. This study aims to analyze the shoreline change of Lake Batur, located in the Batur UNESCO Global Geopark (BUGG), during the period of 2007 – 2018 and design a prediction for 2030. Understanding the shoreline change is very important for lake management and planning. Shoreline changes were analyzed in Geographic Information System (GIS) application. The data were obtained based on Remote Sensing (RS) data, Landsat ETM+ imagery on September 24, 2007, and Landsat OLI imagery on October 24th, 2018. Predictions of lake shoreline in 2030 result from modelling by integrating ASTER-GDEM V2 data, lake water level data for 2007–2018, annual average rainfall data for 2007–2018, and bathymetry data for 2013 and 2015. The results of the satellite imagery analysis show that there has been a change in the length of the shoreline, which has increased from 20.47 km in 2007 to 21.28 km (3.96%) in 2018. The lake surface area changed from 15.34 km2 in 2007 to 16.16 km2 in 2018 (5.35%). The prediction of lake shoreline changes in 2030 showed that Lake Batur will increase to 26.90 km (26.41%), and lake surface area is predicted to increase by 17.67 km2 (9.34%) from 2018. This is because of the morphological change of Lake Bottom.

Assessing water quality at Jakkur Lake: a comparative analysis of Landsat-8 OLI and Sentinel-2 Satellite Imagery

Jakkur lake is one of the biggest lakes of Bangalore and is located on the northern side of the city. It derives its name from the name of the locality, Jakkur. The waterbody area is about 160 acres and length is about 1.5 km with several islands. Because of urbanization and increase in population, the catchment location is highly encroached. The amount of the stormwater achieving the lake has decreased making the streams dry. Remote sensing is a tool used for monitoring water bodies to present better spatial and temporal resolution. Remote sensing is also used for continuous monitoring and assessment of water bodies. Landsat-8 OLI and Sentinel-2 are recently launched multispectral sensors that have the potential to be used for the effective monitoring of the Jakkur Lake's water quality. Landsat-8 OLI and Sentinel-2 are similar in many approaches. Sentinel-2, however, has higher spatial resolution. The aim of this research was to find out if Sentinel-2 produces better results than Landsat-8 OLI in testing water quality parameters in the Jakkur lake. The parameters of interest were chlorophyll_a and turbidity. The results show that Sentinel-2 generates better results. For Sentinel-2, the R2 values for chlorophyll_a and turbidity were 0.975 and 0.845 respectively whilst for Landsat-8, the chlorophyll_a and turbidity R2 values were 0.911 and 0.747 respectively.

Remote monitoring of water clarity in coastal oceans of the Guangdong-Hong Kong-Macao Greater Bay Area, China based on machine learning

The development of the Guangdong-Hong Kong-Macao Greater Bay Area (GBA), one of the most developed and densely populated regions in China, has posed an increasing threat to the health of the water environment in adjacent coastal oceans. However, the spatiotemporal variations of water clarity in the coastal oceans of the GBA (COGBA) have not been well-documented. Therefore, this study aims to develop a remote sensing retrieval algorithm for assessing water clarity, represented by Secchi disk depth (ZSD), based on Landsat-8 OLI imagery for the COGBA using a machine learning method. Cross-validation demonstrated that the algorithm performed exceptionally well, with an R2 value greater than 0.8. By applying the algorithm to 263 Landsat-8 OLI images, we obtained a time series of ZSD for the period of 2013–2021 over the COGBA. Results indicate that the Pearl River Estuary (PRE) exhibited the highest turbidity, followed by the Daya Bay and the Mirs Bay. Generally, ZSD increased from the northwest to the southeast of the COGBA. Seasonal, interannual, and long-term variations in ZSD were also observed with long-term increases in the PRE and the Daya Bay. The interannual variations in ZSD during fall were primarily regulated by different factors in each region. In the PRE, the negative effects of sediment discharge and wind speed played a significant role. In Mirs Bay, wind speed and sediment discharge had a negative impact on ZSD. In Daya Bay, precipitation and wind speed were the key factors influencing ZSD. The development and findings of our algorithm contribute to the protection and management of the water environment in the COGBA, facilitating effective governance measures.

Comparison of Mangrove Index (MI) and Normalized Difference Vegetation Index (NDVI) for the detection of degraded mangroves in Alas Purwo Banyuwangi and Segara Anakan Cilacap, Indonesia

Deterioration of the environmental quality of mangrove ecosystem must be identified as early as possible for management and threat mitigation. Degraded mangrove areas have to be distinguished from healthy ones. Low value of Mangrove Index (MI) corresponded to the degraded mangrove area where dominated by two species (Derris trifoliata and Acanthus ilicifolius), which are indicators of mangrove degradation at Segara Anakan mangrove area. The Normalized Difference Vegetation Index (NDVI) failed to detect the degaradation due to existing of these species by detecting as high index value. The study was designed to (1) test the performance of MI at Alas Purwo mangrove area which categorized as a non-degraded mangrove forest, (2) compare the spatial pattern of MI images with that of NDVI images, and (3) further investigate the ability of MI to detect areas of degraded mangrove when it occurs in only small areas. We used Landsat-8 OLI imagery because it includes the short-wave infrared (SWIR) bands required to generate MI. Both direct visual comparison and correlation analysis of the resulting maps were conducted to evaluate the ability of these indices to detect degraded and non-degraded areas of mangroves. In general, both indices detected a similar pattern, which is contrary to the results from Segara Anakan images. However, a difference value between MI and NDVI of >0.2 indicated the presence of small areas of vegetation with a spectral reflectance similar to that of species indicating degradation at Segara Anakan. This finding was confirmed by field observation of mangrove species. A difference of >0.2 between MI and NDVI indicates mangrove degradation for small areas. This study provides a new tool to determine the condition of mangrove forest remotely, allowing for rapid and targeted management action to tackle degradation.

A study of urban heat island effects using remote sensing and GIS techniques in Kancheepuram, Tamil Nadu, India

Aspects of ecological issues arising from rapid urbanization are investigated in this paper using Land Surface Temperature (LST) in terms of radiation budgets studied by heat conduct balance using a geospatial approach. It uses the data from USGS Landsat ETM+ 4–5 and Landsat-8 OLI imagery acquired on February 6, 1988 and October 14, 2021. The geospatial technique examines the effects of changes in Land Use/Land Cover (LSLC) in terms of NDVI classification on the distribution of surface temperature. The remote sensing approach identifies changes in land use, their influence on surface temperature, and variations in average LST caused by these hotspots. This study examines the distribution of LST in Kancheepuram district, Tamil Nadu, India. The LSTs ranged from 14 °C to 31 °C in 1985, from 25 °C to 39 °C in 2005, and from 31 °C to 47 °C in 2021. The results show a 152% increase in the mean LST in 4 decades. The LST high values were observed in urban and industrial regions with buildings, impervious sidewalks, and sparse vegetation. The intensity at hotspots was highest in urbanized and sparsely vegetated regions. The low Urban Heat Island (UHI) effect was measured in vegetated areas. Industrial areas and parking lots were commonly classified as high-intensity hotspots. The elevated temperatures are attributable to the materials used in the construction at these locations. Further research is needed to better understand the reflectance of thermal radiations by certain materials used in urban regions and how to minimize temperatures in such locations.

Vegetation structural composition mapping of a complex landscape using forest cover density transformation and random decision forest classifier: a comparison

Forest cover density (FCD) transformation and random decision forest (RDF) classification have been widely used for vegetation mapping. Nevertheless, a comparison of their capabilities in complex tropical landscapes is still rarely carried out. This study compared the two methods using Landsat-8 OLI imagery which includes the blue up to thermal bands for vegetation structural composition mapping in a complex landscape of Central Java, Indonesia. We used the FCD transformation with six indices to generate 11 classes, while the RDF classified the same 11 classes based on training areas and used a random process involving various number of splits and trees. The results showed that the FCD transformation achieved 69.32% accuracy, while the RDF was able to classify the 11 classes with various accuracies depending on the parameter setting, i.e. from 70.76% to 75.19%. Regarding the obtained accuracies, problems associated with the terrain and vegetation characteristics have been discussed for further recommendation.

Geo-spatial modelling of carbon stock assessment of date palm at different age stages: An integrated approach of fieldwork, remote sensing and GIS

The aim of this study is to develop a geospatial model for the assessment and mapping of carbon stock (CS) of date palms (DP) in Abu Dhabi, at three age stages: mature, medium and young. The approach relied on the use of correlated remote sensing predictors tested and derived from Landsat-8 imagery over the study period to estimate CS. Multiple predictors including single bands, vegetation indices (VIs) and their combination were used in linear and non-linear regression models to identify the best performing predictors and regression models for DP. Mature DP, older than 10 years, attained the most significant correlation when using a second-order polynomial model with the tasseled cap for wetness as a predictor, yielding R² of 0.7643. For non-mature DP, less than 10 years, the exponential regression model using renormalized difference vegetation index (RDVI) as a predictor provided the most significant correlation for AGB estimation with R2 of 0.4987. The study demonstrated that moderate-resolution Landsat-8 OLI imagery has the potential to estimate the CS in DP of drylands at different age stages. This is the first attempt to estimate CS in DP using geospatial technologies with minimal fieldwork hence, expanding our approach to other remote and less-studied drylands.

INVESTIGATING THE IMPACT OF PAN SHARPENING ON THE ACCURACY OF LAND COVER MAPPING IN LANDSAT OLI IMAGERY

Pan Sharpening is normally applied to sharpen a multispectral image with low resolution by using a panchromatic image with a higher resolution, to generate a high resolution multispectral image. The present study aims at assessing the power of Pan Sharpening on improvement of the accuracy of image classification and land cover mapping in Landsat 8 OLI imagery. In this respect, different Pan Sharpening algorithms including Brovey, Gram-Schmidt, NNDiffuse, and Principal Components were applied to merge the Landsat OLI panchromatic band (15 m) with the Landsat OLI multispectral: visible and infrared bands (30 m), to generate a new multispectral image with a higher spatial resolution (15 m). Subsequently, the support vector machine approach was utilized to classify the original Landsat and resulting Pan Sharpened images to generate land cover maps of the study area. The outcomes were then compared through the generation of confusion matrix and calculation of kappa coefficient and overall accuracy. The results indicated superiority of NNDiffuse algorithm in Pan Sharpening and improvement of classification accuracy in Landsat OLI imagery, with an overall accuracy and kappa coefficient of about 98.66% and 0.98, respectively. Furthermore, the result showed that the Gram-Schmidt and Principal Components algorithms also slightly improved the accuracy of image classification compared to original Landsat image. The study concluded that image Pan Sharpening is useful to improve the accuracy of image classification in Landsat OLI imagery, depending on the Pan Sharpening algorithm used for this purpose.

Mapping of Limestone Potential Using Landsat 8 Satellite Imageryin Some Areasof Timpeh

Limestone potential is important information that can be obtained from remote sensing data which has advantages and speed in processing results. Remote sensing is a technology that can overcome the problemof measuring data for fast and accurate information. This research was carried out in some areas of the Timpeh sub-district,andDharmasraya districtusing Landsat 8-OLI imagery with the aimof1) identifying the potential of limestone using the Band Ratio method. 2) How to apply remote sensing in mapping the potential of limestoneusing Landsat 8 Oli imagery.
 This research was carried out in several stages, namely Pre Processing which included radiometric correction and atmospheric correction, image cropping according to the research area, and processing whichincluded making geological maps, making landform maps, making maps of river flow patterns and vegetationindex maps and limestone identification using the RGB band ratio method (5/4;6/3;4/2).
 The results of field identification in potential limestone areas, where the RGB (Red Green Blue)composite of the band ratio 5/4;6/3;4/2 shows that the presence of limestone is characterized by the appearanceof greenish-brown colored objects. The average pixel value for limestone with a band ratio of 5/4 is 2.475, for a6/3 ratio is 1.275 and for a 4/3 ratio is 0.788. In this study, the potential area of limestone in the research areawasfound,whichwas approximately 2352,14564 ha.

The Effect of Temporal Resolution of Climatic Factors on Agriculture Degradation in Southern Baghdad by Applying Remote Sensing Data

The climate changes had been recognized as one of the major factors responsible for land degradation, which has a significant impact on diverse aspects. The present study aims to estimate how the climate change can influence land degradation in the south areas of Baghdad province (Al-Rasheed, Al-Mahmudiyah, Al-Yusufiyah, Al-Madaen, and Al-Latifiyah). The Satellite Landsat-8 OLI and satellite Landsat-5 TM sensor imagery were used to extent land degradation for the period (2010-2019). ArcGIS V.10.4 was applied to manage and analysis the satellite image dataset, including the use of climate factors data from the European Center for Climate Forecasts (ECMWF) by reanalyzes and extraction datasets. To achieve work objectives, many ground data were collected, including the temperature, rain precipitation, evaporation, and relative humidity from 30 meteorological monitoring stations. These data help us to utilize the interpolation methods for the extraction process of contour lines maps, to be scientific indicators of the relationship between climatic factors and satellite images classifications, involving the spectral indicators of the vegetation cover and water bodies. The results showed the agriculture degradation through the decreasing of vegetation cover rate from 56.57% in (2010) to 43.43% in )2019 (. This deterioration is thought to be related to climate changes with other factors such as water shortage that was 0.52 and 0.44, respectively, the greatest temperature reading was (24.57), the greatest precipitation was (0.21), the greatest relative humidity was (60.73), and vapor rate (-0.2) for the studied period.

IDENTIFICATION OF LAND COVER OF SURFACE TEMPERATURE IN KOTA PARIAMAN USING LANDSAT IMAGERY 8-OLI

Remote sensing is a technology that can overcome data measurement problems for fast and accurate information. This research was carried out in Pariaman City using Landsat 8-OLI imagery with the aim of 1) knowing the identification of land cover based on the classification SNI:7645 (2014) based on Landsat 8-OLI imagery in the Pariaman City area in 2020 2) knowing the identification of surface temperature in the Pariaman City area in 2020. The research was conducted in serval stages, namely pre-processing of image data, selecting band combinations, cutting area studies, image classification, and testing accuracy. At surface temperature using Landsat 8-OLI imagery in 2020, surface temperature values ​​are obtained from the results of thermal band processing. The results of the classification of Landsat 8-OLI images in Pariaman City produce 5 land cover classes, namely water bodies, rice fields, settlements, mixed gardens, and shrubs. This research conducted a classification accuracy test using a confusion matrix accuracy test table. Land cover supervised maximum likelihood method the overall accuracy value is 86.66 %. The results of the surface temperature value in Pariaman City in 2020 obtained the highest temperature value of 30ºC and the lowest surface temperature of 23ºC.

Mapping plastic-covered greenhouse farming areas using high-resolution PlanetScope and RapidEye imagery: studies from Loukkos perimeter (Morocco) and Dalat City (Vietnam).

The proliferation of plastic-covered greenhouse (PCG) farming has resulted in high horticulture crop yields worldwide during the last few decades. A proper and cost-effective PCG monitoring method is necessary for maintaining sustainable horticulture and high-quality agricultural production with less plastic pollution. Remote sensing applications for mapping PCG have received great attention from the scientific community in recent years. In this paper, a comparative study was carried out in two plastic-covered greenhouse areas in Loukkos perimeter in Morocco and Dalat City in Vietnam to test PCG mapping accuracy of high spatial resolution RapidEye and PlanetScope satellite data and to understand the differences in PCG mapping quality due to topographic effects. Medium-resolution Landsat-8 OLI and Sentinel-2 MSI imagery were also applied. Moreover, two classification algorithms-retrogressive plastic greenhouse index (RPGI) and a supervised classification algorithm using random forest (RF)-were used for mapping PCG. The findings reveal that RF outperforms RPGI. Overall, the mapping accuracy achieved exceeded 90% in both study areas, except for the RPGI method using Landsat-8 data (PCG mapping accuracy using Landsat data varied between 87.4 and 89%). Furthermore, PCGs were better detected by PlanetScope data than by RapidEye imagery due to the differences in the spectral range. Better performance in Loukkos perimeter can be explained by the study area's topography; Dalat City and surrounding areas are situated in mountainous terrain. The results obtained from this study indicate that spectral indices can be used as a cost-effective tool for mapping PCG under cloud-free conditions. PCG mapping using RF classifiers resulted in accurate PCG mapping without topographic factors' influence.

GIS investigation of the fire history of Jonkershoek Nature Reserve

Fire regimes have the potential to disturb ecological aspects of a landscape and/or contribute to the maintenance of the biological diversity. Thus, a gauge of the impact of planned and unplanned fire regimes is vital to South Africa’s national reserves. The Jonkershoek Nature Reserve in the Western Cape is characterized by the occurrence of indigenous Fynbos and Afromontane Forest vegetation. Geographical Information Systems (GIS) and Remote Sensing (RS) can aid the management and preservation of indigenous vegetational species. This study used knowledge of the ecological conditions of the Reserve, historical fire data, Landsat TM and Landsat OLI imagery, and geospatial analysis to investigate the impact of the fire regimes in the Reserve. Image classification was carried out from 2005 to 2015 to determine the burn patterns, with the process being aided by the fire regime history from 1970 to 2015. Ordinary Least Squares (OLS) analysis was carried out to determine how abiotic factors, such as elevation, slope and aspect, impact fires in the Reserve. The assessment of fires included the ascertainment of their location, coverage, and frequency, the Normalised Burn Ratio (NBR), the differenced Normalised Burn Ratio (dNBR) and the Normalised Difference Vegetation Index (NDVI). There were 39 fires recorded in the Jonkershoek Nature Reserve from 1970 to 2015. The largest fire events were recorded in 1999 (26503.6 ha.) and 2015 (8363.0 ha.). The lowest area of fire impact recorded occurred in the years 2010 (0.15ha.), 1973 (1.1 ha.) and 1987 (3.1 ha.). With an overall classification accuracy of 94.17%, the Landsat OLI imagery performed better with an overall classification accuracy of 94.17% than the Landsat TM at 75.83%. The OLS regression showed that fire severity was positively correlated to NDVI and elevation. This may suggest that regions of healthy vegetation at any altitude may be susceptible to burnings if there is sufficient vegetation to fuel a fire. The OLS was negatively correlated to slope and aspect. This may impact fire risk as steeper slopes may have vegetation growing in their fire shadow.

Classifying Forest Types over a Mountainous Area in Southwest China with Landsat Data Composites and Multiple Environmental Factors

Accurate information about forest type and distribution is critical for many scientific applications. It is possible to make a forest type map from the satellite data in a cost effective way. However, forest type mapping over a large and mountainous geographic area is still challenging, due to complex forest type compositions, spectral similarity among various forest types, poor quality images with clouds or cloud shadows and difficulties in managing and processing large amount data. Based on the Google Earth Engine (GEE) cloud platform, a method of forest types mapping using Landsat-8 OLI imagery and multiple environmental factors was developed and tested within Yunnan Province (about 390,000 km2) of China. The proposed approach employed a pixel-based seasonal image compositing method to produce two types of seasonal composite images, i.e., four 7-spectral-band composite images and four 5-VI-band composite images associated in spring, summer, autumn, and winter. Then, single-season feature bands and multi-seasonal feature bands were combined with the feature bands of topography, temperature, and precipitation, respectively, and resulting in 17 feature combinations. Finally, using a random forest (RF) classifier, 17 feature combinations were separately experimented to classify the forest type over the study area. The study area was firstly classified into the forest and the non-forest, and then the forest was sub-classified into five forest types (evergreen needleleaf forest, deciduous needleleaf forest, evergreen broadleaf forest, deciduous broadleaf forest, and mixed forest). The results showed that the pixel-based multi-seasonal median composite can produce a cloud-free image for the entire region and is suitable for forest type mapping. Compared with a single-season composite, a multi-seasonal composite can distinguish different forest types more effectively. The environmental factors also improve the accuracy of forest type mapping. With the ground survey samples as reference values, the classification performance of 17 feature combinations was compared, and the optimal feature combination was found out. For the optimal feature combination, its overall accuracy of the forest/non-forest cover map and the forest type map reached 97.57% (Kappa = 0.950) and 70.30% (Kappa = 0.628), respectively. The proposed approach has demonstrated strong potential of high classification accuracy and convenient calculation when mapping forest types over a national or global scale, and its product of 30 m resolution forest type map is capable of contributing to forest resource management.

AGTOC: A novel approach to winter wheat mapping by automatic generation of training samples and one-class classification on Google Earth Engine

Accurate and timely acquisition of crop spatial distribution is a prerequisite for growth monitoring and yield forecasting. Currently, the automatic acquisition of crop distribution at large scales is still a challenge due to the time-consuming processing of remotely sensed imagery and manual collection of sufficient training samples. Although the advent of cloud computing platforms has proved to improve the efficiency and automation of crop type classification, how to obtain sufficient training samples in an efficient and cost-effective way remains unclear. In this research, we developed a new approach integrating the automatic generation of training samples and one-class machine learning classification (AGTOC) for mapping winter wheat over Jiangsu Province, China on Google Earth Engine (GEE). After extracting spatial objects from Sentinel-2 imagery in the season of 2017–2018, this method performed recognition of winter wheat objects based on the unique phenology and spectral features of winter wheat. Then the generated winter wheat objects were further refined and regarded as training samples for provincial winter wheat classification with one-class support vector machine (OCSVM). Furthermore, the transferability of AGTOC was evaluated by applying the classification approach to different seasons (2016–2017 & 2019–2020) and a different sensor (Landsat-8 OLI). According to independent ground truth data, the winter wheat mapping with AGTOC achieved an overall accuracy (OA) of 92.61%. When compared with agricultural census data, the winter wheat area accounted for 99% and 90% of the variability at the municipal and county levels. Furthermore, the OA achieved 88.94% and 90.17% while transferring the AGTOC from 2017–2018 to 2016–2017 and 2019–2020. The transferability of the AGTOC model to Landsat-8 OLI imagery of the same season yielded an OA of 85.98%. These results demonstrated that AGTOC exhibited high efficiency and accuracy across the province, different seasons and sensors without the need of extensive field visits for training sample collection. This proposed approach has great potential in the automatic mapping of winter wheat on GEE at country or global levels.

Mapping desertification degree and assessing its severity in Al-Ahsa Oasis, Saudi Arabia, using remote sensing-based indicators

Desertification in Saudi Arabia is cause for environmental and socio-economic concern, as it has severe and problematic impacts. This study aimed to map desertification and the degree of its severity over the date palms-dominated Al-Ahsa Oasis in the Eastern Province of Saudi Arabia using Landsat-8 OLI imagery acquired in December 2019. Based on linear regression analysis, correlations between the normalized difference vegetation index (NDVI) and albedo (α), tasseled cap wetness (TCW), brightness (TCB), and greenness (TCG) were obtained. These indicators were generated by employing the tasseled cap transformation method (TCT). The results revealed a strong negative association between TCW and TCB (r = − 0.91). In contrast, NDVI–α exhibited a weak negative relationship (r = − 0.53) as a desertification identifier. Thus, the desertification degree index (DDI) was applied to the TCW–TCB relationship. The findings indicated that 83% of the oasis exhibited desertification, which was further classified based on the degree of severity into 14.2% extreme, 19.3% severe, 23.9% moderate, and 25.6% low. Overall, the study illustrates the capability of remote sensing data and techniques in mapping and assessing the degree of desertification and its severity by applying the DDI index.

Using an explicit multi-feature decision tree for wetland information extraction in Qomolangma national nature reserve

ABSTRACT The Qomolangma National Nature Reserve (QNNR) has complex geomorphological characteristics of the high altitude Tibetan Plateau and aunique plateau wetland environment. In order to extract the wetland information effectively and accurately, this research utilizes an explicit decision tree (DT) classification strategy that incorporates multiple feature nodes by using the Landsat-8 OLI imagery as the main data source to interpret the spectral signatures and spatial distribution of the wetlands, and to construct acomprehensive DT classification process for wetland information extraction in QNNR. The multiple feature nodes include snow cover, non-wetland distribution, normalized difference vegetation index (NDVI), normalized difference water index (NDWI), digital elevation model (DEM), land surface temperature (LST) and slope. This strategy produces amore accurate wetland classification in QNNR where there are various interfering factors such as snow, mountain shadows, mineralized regions, grassland and woodland. The method is straightforward and easy to implement. Its classification result has an overall accuracy of 92.75%. This method could be applied to the plateaus with vertical-step ecosystems.