Existing Land Cover Products Research Articles

Mapping Quaking Aspen Using Seasonal Sentinel-1 and Sentinel-2 Composite Imagery across the Southern Rockies, USA

Quaking aspen is an important deciduous tree species across interior western U.S. forests. Existing maps of aspen distribution are based on Landsat imagery and often miss small stands (<0.09 ha or 30 m2), which rapidly regrow when managed or following disturbance. In this study, we present methods for deriving a new regional map of aspen forests using one year of Sentinel-1 (S1) and Sentinel-2 (S2) imagery in Google Earth Engine. Using observed annual phenology of aspen across the Southern Rockies and leveraging the frequent temporal resolution of S1 and S2, ecologically relevant seasonal imagery composites were developed. We derived spectral indices and radar textural features targeting the canopy structure, moisture, and chlorophyll content. Using spatial block cross-validation and Random Forests, we assessed the accuracy of different scenarios and selected the best-performing set of features for classification. Comparisons were then made with existing landcover products across the study region. The resulting map improves on existing products in both accuracy (0.93 average F1-score) and detection of smaller forest patches. These methods enable accurate mapping at spatial and temporal scales relevant to forest management for one of the most widely distributed tree species in North America.

Hybrid Global Annual 1-km IGBP Land Cover Maps for the Period 2000–2020

Land cover products provide critical information for monitoring and analyzing land surface changes. However, notable disagreement and incompatible classification systems among existing land cover products bring challenges in using them. Here, we developed a hierarchical International Geosphere-Biosphere Programme (IGBP) classification system and integrated four widely used land cover products (i.e., MODIS-IGBP, ESA-CCI, GlobeLand30, and GLC_FCS30) based on their accuracy against a collection of global reference samples. We generated a hybrid global annual land cover product (HYBMAP) with ~1 km (1/120°, 30″) spatial resolution from 2000 to 2020. The HYBMAP integrates information from the four products of high- and medium-resolution and reduces the disagreement between them by up to 20.1%. The overall accuracy of the HYBMAP is 75.5%, which is higher than the best of the four products (MODIS-IGBP, 70.9%). HYBMAP also integrates the temporal change information from the four products and identifies a faster growth of built-up lands. The HYBMAP provides more consistent and reliable global land cover time series data for global change research. It is free to access at https://doi.org/10.5281/zenodo.10488191 .

Regional Accuracy Assessment of 30-Meter GLC_FCS30, GlobeLand30, and CLCD Products: A Case Study in Xinjiang Area

With the development of remote sensing technology, a number of fine-resolution (30-m) global/national land cover (LC) products have been developed. However, accuracy assessments for the developed LC products are commonly conducted at global and national scales. Due to the limited availability of representative validation observations and reference data, knowledge relating to the accuracy and applicability of existing LC products on a regional scale is limited. Since Xinjiang, China, exhibits diverse surface cover and fragmented urban landscapes, existing LC products generally have high classification uncertainty in this region. This makes Xinjiang suitable for assessing the accuracy and consistency of exiting fine-resolution land cover products. In order to improve knowledge of the accuracy of existing fine-resolution LC products at the regional scale, Xinjiang province was selected as the case area. First, we employed an equal-area stratified random sampling approach with climate, population density, and landscape heterogeneity information as constraints, along with the hexagonal discrete global grid system (HDGGS) as basic sampling grids to develop a high-density land cover validation dataset for Xinjiang (HDLV-XJ) in 2020. This is the first publicly available regionally high-density validation dataset that can support analysis at a regional scale, comprising a total of 20,932 validation samples. Then, based on the generated HDLV-XJ dataset, the accuracies and consistency among three widely used 30-m LC products, GLC_FCS30, GlobeLand30, and CLCD, were quantitatively evaluated. The results indicated that the CLC_FCS30 exhibited the highest overall accuracy (88.10%) in Xinjiang, followed by GlobeLand30 (with an overall accuracy of 83.58%) and CLCD (81.57%). Moreover, through a comprehensive analysis of the relationship between different environmental conditions and land cover product performance, we found that GlobeLand30 performed best in regions with high landscape fragmentation, while GLC_FCS30 stood out as the most outstanding product in areas with uneven proportions of land cover types. Our study provides a novel insight into the suitability of these three widely-used LC products under various environmental conditions. The findings and dataset can provide valuable insights for the application of existing LC products in different environment conditions, offering insights into their accuracies and limitations.

A super-ensemble approach to map land cover types with high resolution over data-sparse African savanna landscapes

Accurate and timely land cover products are critical inputs for landscape planning, and provide key information for biodiversity conservation and food security. However, poor mapping quality and low resolution are considerable issues in existing land cover maps over the African savanna, where land use is complex and changing rapidly, and necessary ground-truth data are sparse and hard to obtain. To overcome this problem, to make optimal use of existing maps, and to minimize manual training data collection, we developed a three-stage ensemble method to make land cover maps. In the first stage, we extracted the consensus of multiple existing land cover products to generate fragmented pixel-wise training labels. In the second stage, we translated pixel-wise training labels to image-wise labels using Random Forest (RF) as a “gap-filling model”, with temporal features extracted from Sentinel-1 time series, raw bands, and vegetation indices derived from PlanetScope basemaps. These image-wise labels were scored and edited by humans and the quality information was used in the next stage. For stage three, we trained a U-Net network based upon these image-wise labels, using Sentinel-1 time series and raw bands of PlanetScope basemaps as image features. Using the information on label quality, a quality-weighted loss function was used in the network to reduce the impact of noise in the training labels. Using Northern Tanzania as a case study, the results demonstrate that ensembles of existing land cover maps provide a useful source of data for developing improved land cover maps over hard-to-classify, data-sparse landscapes. The Random Forest “gap-filling model” had an overall accuracy of 80.26% on our independent test dataset with 7 classes. The final U-Net model had an overall accuracy of 83.57%. This approach can be readily applied to other regions and extents (e.g., regional, global) and other data sources (e.g., Sentinel-2).

Improved forest cover mapping by harmonizing multiple land cover products over China

ABSTRACT Fine resolution land cover products are becoming increasingly more available for many regions. These products, however, may not meet the quality requirements of many applications. This study provides an approach for improving land cover mapping by leveraging existing products and clear view Landsat composites. Assessments using independent reference datasets revealed that the CAF-LC30 2020 product derived using this approach over China was more accurate than four existing land cover products. Its overall accuracy with field observations was 2.94% to 10.28% higher than those of the four existing land cover products in northeast China and was 2.10% to 8.18% better across China. It provided a more accurate representation of the land cover types in many regions where the existing land cover products had large classification errors. Forest areas calculated using the CAF-LC30 2020 for the 31 provinces and autonomous regions and municipalities (PARM) in mainland China were better correlated with those reported by the most recent National Forest Inventory (NFI) survey than areas calculated using the other four existing land cover products. Therefore, the CAF-LC30 2020 product should be a better alternative for understanding China’s forests in 2020 than the other four existing land cover products.

Time-series land cover mapping and urban expansion analysis using OpenStreetMap data and remote sensing big data: A case study of Guangdong-Hong Kong-Macao Greater Bay Area, China

• Optimized training samples from OpenStreetMap facilitate large-scale land cover mapping. • Sample optimization and transfer, and consistency modification strategies improve mapping results. • The proposed workflow achieves satisfactory performance and is transferable to other regions. Accurate land cover mapping provides important scientific support for ecological environment protection and sustainable urban development. However, given the high expense of acquiring training samples and the difficulty of fully utilizing remote sensing big data, large-scale time-series land cover mapping remains to be a challenge. To address the issue, we proposed a novel time-series large-scale mapping approach that obtains high-quality training samples from OpenStreetMap volunteered data and transfers them for land cover mapping in historical years. Relying on the data archived on the Google Earth Engine platform, we constructed a discriminating feature set that contains spectrum, texture, and backscatter coefficient, among others. Taking the Guangdong-Hong Kong-Macao Greater Bay Area as a case area, the annual land cover maps after spatio-temporal consistency modification from 1986 to 2021 were obtained. The validation samples proved that the derived land cover classification results have high accuracy. We verified the reliability and superiority of the proposed approach by comparing our results with existing land cover products. The proposed mapping approach owns great transferability, and the mapping results provide a valid decision-making basis for urban planning.

Comparing CNNs and Random Forests for Landsat Image Segmentation Trained on a Large Proxy Land Cover Dataset

Land cover mapping from satellite images has progressed from visual and statistical approaches to Random Forests (RFs) and, more recently, advanced image recognition techniques such as convolutional neural networks (CNNs). CNNs have a conceptual benefit over RFs in recognising spatial feature context, but potentially at the cost of reduced spatial detail. We tested the use of CNNs for improved land cover mapping based on Landsat data, compared with RFs, for a study area of approximately 500 km × 500 km in southeastern Australia. Landsat 8 geomedian composite surface reflectances were available for 2018. Label data were a simple nine-member land cover classification derived from reference land use mapping (Catchment Scale Land Use of Australia—CLUM), and further enhanced by using custom forest extent mapping (Forests of Australia). Experiments were undertaken testing U-Net CNN for segmentation of Landsat 8 geomedian imagery to determine the optimal combination of input Landsat 8 bands. The results were compared with those from a simple autoencoder as well as an RF model. Segmentation test results for the best performing U-Net CNN models produced an overall accuracy of 79% and weighted-mean F1 score of 77% (9 band input) or 76% (6 band input) for a simple nine-member land cover classification, compared with 73% and 68% (6 band input), respectively, for the best RF model. We conclude that U-Net CNN models can generate annual land cover maps with good accuracy from proxy training data, and can also be used for quality control or improvement of existing land cover products.

History of Land Cover Change on Santa Cruz Island, Galapagos

Islands are particularly vulnerable to the effects of land cover change due to their limited size and remoteness. This study analyzes vegetation cover change in the agricultural area of Santa Cruz (Galapagos Archipelago) between 1961 and 2018. To reconstruct multitemporal land cover change from existing land cover products, a multisource data integration procedure was followed to reduce imprecision and inconsistencies that may result from the comparison of heterogeneous datasets. The conversion of native forests and grasslands into agricultural land was the principal land cover change in the non-protected area. In 1961, about 94% of the non-protected area was still covered by native vegetation, whereas this had decreased to only 7% in 2018. Most of the agricultural expansion took place in the 1960s and 1970s, and it created an anthropogenic landscape where 67% of the area is covered by agricultural land and 26% by invasive species. Early clearance of native vegetation took place in the more accessible—less rugged—areas with deeper-than-average and well-drained soils. The first wave of settlement consisted of large and isolated farmsteads, with 19% of the farms being larger than 100 ha and specializing in diary and meat production. Over the period of 1961–1987, the number of farms doubled from less than 100 to more than 200, while the average farm size decreased from 90 to 60 ha/farmstead. Due to labor constraints in the agricultural sector, these farms opted for less labor-intensive activities such as livestock farming. New farms (popping up in the 1990s and 2000s) are generally small in size, with <5 ha per farmstead, and settled in areas with less favorable biophysical conditions and lower accessibility to markets. From the 1990s onwards, the surge of alternative income opportunities in the tourism and travel-related sector reduced pressure on the natural resources in the non-protected area.

Testing Accuracy of Land Cover Classification Algorithms in the Qilian Mountains Based on GEE Cloud Platform

The Qilian Mountains (QLM) are an important ecological barrier in western China. High-precision land cover data products are the basic data for accurately detecting and evaluating the ecological service functions of the QLM. In order to study the land cover in the QLM and performance of different remote sensing classification algorithms for land cover mapping based on the Google Earth Engine (GEE) cloud platform, the higher spatial resolution remote sensing images of Sentinel-1 and Sentinel-2; digital elevation data; and three remote sensing classification algorithms, including the support vector machine (SVM), the classification regression tree (CART), and the random forest (RF) algorithms, were used to perform supervised classification of Sentinel-2 images of the QLM. Furthermore, the results obtained from the classification process were compared and analyzed by using different remote sensing classification algorithms and feature-variable combinations. The results indicated that: (1) the accuracy of the classification results acquired by using different remote sensing classification algorithms were different, and the RF had the highest classification accuracy, followed by the CART and the SVM; (2) the different feature variable combinations had different effects on the overall accuracy (OA) of the classification results and the performance of the identification and classification of the different land cover types; and (3) compared with the existing land cover products for the QLM, the land cover maps obtained in this study had a higher spatial resolution and overall accuracy.

An Adaptive-Parameter Pixel Unmixing Method for Mapping Evergreen Forest Fractions Based on Time-Series NDVI: A Case Study of Southern China

Spectral unmixing remains the most popular method for estimating the composition of mixed pixels. However, the spectral-based unmixing method cannot easily distinguish vegetation with similar spectral characteristics (e.g., different forest tree species). Furthermore, in large areas with significant heterogeneity, extracting a large number of pure endmember samples is challenging. Here, we implement a fractional evergreen forest cover-self-adaptive parameter (FEVC-SAP) approach to measure FEVC at the regional scale from continuous intra-year time-series normalized difference vegetation index (NDVI) values derived from moderate resolution imaging spectroradiometer (MODIS) imagery acquired over southern China, an area with a complex mixture of temperate, subtropical, and tropical climates containing evergreen and deciduous forests. Considering the cover of evergreen forest as a fraction of total forest (evergreen forest plus non-evergreen forest), the dimidiate pixel model combined with an index of evergreen forest phenological characteristics (NDVIann-min: intra-annual minimum NDVI value) was used to distinguish between evergreen and non-evergreen forests within a pixel. Due to spatial heterogeneity, the optimal model parameters differ among regions. By dividing the study area into grids, our method converts image spectral information into gray level information and uses the Otsu threshold segmentation method to simulate the appropriate parameters for each grid for adaptive acquisition of FEVC parameters. Mapping accuracy was assessed at the pixel and sub-pixel scales. At the pixel scale, a confusion matrix was constructed with higher overall accuracy (87.5%) of evergreen forest classification than existing land cover products, including GLC 30 and MOD12. At the sub-pixel scale, a strong linear correlation was found between the cover fraction predicted by our method and the reference cover fraction obtained from GF-1 images (R2 = 0.86). Compared to other methods, the FEVC-SAP had a lower estimation deviation (root mean square error = 8.6%). Moreover, the proposed method had greater estimation accuracy in densely than sparsely forested areas. Our results highlight the utility of the adaptive-parameter linear unmixing model for quantitative evaluation of the coverage of evergreen forest and other vegetation types at large scales.

Growing Spatial Overlap Between Dam-Related Flooding, Cropland and Domestic Water Points: A Water–Energy–Food Nexus Management Challenge in Malawi and Ghana

In sub-Saharan Africa, land cover change, expansion of hydropower infrastructure, and increased flooding complicate country-level efforts to meet the Sustainable Development Goal target concerning access to safe water. The Water, Energy and Food (WEF) nexus approach recognises that addressing these complex challenges requires cross-sectoral analyses at multiple scales. Building on such an approach, our study examined the interrelationships between land cover change, dam-related flooding and access to safe water via a national-level spatial analysis with local case studies in Malawi and Ghana. Our assessment of the water–food interactions found that areas of overlap between water points and cropland increased from 2000 to 2020 for both countries at national scale, but overlap extent varied greatly depending on the land cover product used. Local-scale exploration of water point installation patterns in Zomba, Malawi confirmed this pattern, highlighting increasing non-governmental funding of borehole installation programmes. Our assessment of water–energy interactions found that flooding mediated by hydropower dams increased for the White Volta Basin in Ghana, thereby increasing inundation of groundwater points. Local-scale focus group discussions revealed flooding resulted in contaminated water sources and high risk of injury or drowning whilst fetching water. Overall, our study highlights how socio-economic drivers are bringing water points, flooding and cropland into closer proximity, requiring flood mitigation measures at water points and agro-chemical management to minimise potential water quality impacts. Given differences between land cover products, we recommend more robust integration of existing land cover products to better monitor these phenomena.

Using Eco-Geographical Zoning Data and Crowdsourcing to Improve the Detection of Spurious Land Cover Changes

To address problems in remote sensing image change detection, this study proposes a method for identifying spurious changes based on an eco-geographical zoning knowledge base and crowdsourced data mining. After preliminary change detection using the super pixel cosegmentation method, eco-geographical zoning is introduced, and the rules of spurious change are collected based on the knowledge of expert interpreters, and from statistics on existing land cover products according to each eco-geographical zone. Uncertain changed patches with a high possibility of spurious change according to the eco-geographical zoning rule were published in the form of a map service on an online platform, and then crowd tagging information on spurious changed patches was collected. The Hyperlink-Induced Topic Search (HITS) algorithm was used to calculate the spurious change degree of changed patches. We selected the northern part of Laos as the experimental area and the Chinese GF-1 Wide Field View (WFV) images for change detection to verify the effectiveness of the method. The results show that the accuracy of change detection improves by 23% after removing the spurious changes. Spurious changes caused by clouds, river water turbidity, spectral differences in cultivated land before and after harvest, and changes in shrubs, grassland, and forest density, can be removed using an eco-geographical zoning knowledge base and crowdsourced data mining methods.

Mapping Rural Settlements from Landsat and Sentinel Time Series by Integrating Pixel- and Object-Based Methods

Rural settlements account for 45% of the world’s population and are targeted places for poverty eradication. However, compared to urban footprints, the distribution of rural settlements is not well characterized in most existing land use and land cover maps because of their patchy and scattered organization and relative stability over time. In this study, we proposed a pixel- and object-based method to map rural settlements by employing spectral-texture-temporal information from Landsat and Sentinel time series. Spectral indices (maximum normalized difference vegetation index (NDVI) and minimum normalized difference built-up index (NDBI composite) and texture indices (vertical transmit and vertical receive (VV) polarization of mean synthetic aperture radar (SAR) composite) were calculated from all available Landsat and Sentinel-1A data from 1 January 2016 to 31 December 2018. These features were then stacked for segmentation to extract potential rural settlement objects. To better differentiate settlements from bare soil, the gradient of annual NDVI maximum (namely, gradient of change, use gradient for simplicity) from 1 January 1987 to 31 December 2018 was used. The rural training samples were selected from global urban footprint (GUF) products with a post filtering process to remove sample noise. Scatter plots between pixel- and object-based values per feature were delineated by t-distribution ellipses to determine the thresholds. Finally, pixel- and object-based thresholds were applied to four features (NDVI, NDBI, VV, gradient) in Google Earth Engine (GEE) to obtain the distribution of rural settlements in eight selected Asian regions. The derived maps of rural settlements showed consistent accuracy, with a producer’s accuracy (PA) of 0.87, user’s accuracy (UA) of 0.93 and overall accuracy (OA) reaching 90% in different landscape conditions, which are better than existing land cover products.

Improved Mapping Results of 10 m Resolution Land Cover Classification in Guangdong, China Using Multisource Remote Sensing Data With Google Earth Engine

Land cover information depicting the complex interactions between human activities and surface change is critically essential for nature conservation, social management, and sustainable development. Recent advances have shown great potentials of remote sensing data in generating high-resolution land cover maps, but it remains unclear how different models, data sources, and inclusive features affect the classification results, which hinders its applications in regional studies requiring more accurate land cover data. Informing these issues, here we developed a robust framework to improve the mapping results of 10 m resolution land cover classification in Guangdong Province, China using thousands of manually collected samples, multisource remote sensing data (Sentinel-1, Sentinel-2, and Luojia-1), machine learning algorithms, and a free cloud-based platform of Google Earth Engine. Results showed that an overall accuracy of 86.12% and a Kappa coefficient of 0.84 could be achieved for land cover classification in Guangdong for 2019. We found that random forest models achieved better performance than classification and regression trees, minimum distance, and support vector machine models. We also found that features derived from Sentinel-1 data and Sentinel-2 spectral indices greatly contributed to the classification process, while the feature of Luojia-1 data was not as much important as other configurations. A comparison between our results and several existing land cover products in terms of classification accuracy and visual interpretation further validated the effectiveness and robustness of the proposed framework. Our experiments and findings not only systematically elucidate the role of classification methods and data sources in deriving more accurate and reliable land cover maps but also provide certain guidelines for future land cover mapping from regional to global scales.

Improved Mapping of Mountain Shrublands Using the Sentinel-2 Red-Edge Band

Shrub encroachment into grassland and rocky habitats is a noticeable land cover change currently underway in temperate mountains and is a matter of concern for the sustainable management of mountain biodiversity. Current land cover products tend to underestimate the extent of mountain shrublands dominated by Ericaceae (Vaccinium spp. (species) and Rhododendron ferrugineum). In addition, mountain shrubs are often confounded with grasslands. Here, we examined the potential of anthocyanin-responsive vegetation indices to provide more accurate maps of mountain shrublands in a mountain range located in the French Alps. We relied on the multi-spectral instrument onboard the Sentinel-2A and 2B satellites and the availability of red-edge bands to calculate a Normalized Anthocyanin Reflectance Index (NARI). We used this index to quantify the autumn accumulation of anthocyanin in canopies dominated by Vaccinium spp. and Rhododendron ferrugineum and compared the effectiveness of NARI to Normalized Difference Vegetation Index (NDVI) as a basis for shrubland mapping. Photointerpretation of high-resolution aerial imagery, intensive field campaigns, and floristic surveys provided complementary data to calibrate and evaluate model performance. The proposed NARI-based model performed better than the NDVI-based model with an area under the curve (AUC) of 0.92 against 0.58. Validation of shrub cover maps based on NARI resulted in a Kappa coefficient of 0.67, which outperformed existing land cover products and resulted in a ten-fold increase in estimated area occupied by Ericaceae-dominated shrublands. We conclude that the Sentinel-2 red-edge band provides novel opportunities to detect seasonal anthocyanin accumulation in plant canopies and discuss the potential of our method to quantify long-term dynamics of shrublands in alpine and arctic contexts.

Rapid assessment of juniper distribution in prairie landscapes of the northern Great Plains

Woody plant species including eastern redcedar (Juniperus virginiana) and Rocky Mountain juniper (Juniperus scopulorum) are expanding throughout the prairie ecosystems of the Great Plains because of fire suppression, land management practices, and climate change. Juniper encroachment threatens native grasslands by altering soil characteristics, limiting herbaceous biomass, hindering native plant regeneration, and reducing rangeland productivity. Existing land cover products do not effectively characterize the distribution of juniper over a range of densities, making it difficult to assess the scale of the problem. We evaluated a method for rapid classification and mapping of juniper using matched filtering with Landsat 8 snow-covered and snow-free winter imagery (January-March), and snow-free spring imagery (April–June) for 2015–2016. We used data from two path/rows (29/30 and 30/30) in southeastern South Dakota and northeastern Nebraska (approximately 23,000 km2). In both path/rows, we found that snow-covered winter images increased contrast between juniper and the image background and resulted in the highest overall classification accuracies of 94.5% and 88.9% for juniper densities above 15%, compared to 91.4% and 85.7% for snow-free winter imagery and 57.8% and 74.1% for growing season imagery. Using winter imagery, we successfully captured pixels containing juniper density above 50% with ≥90% detection probability. However, the true positive rate dropped to less than 50% once juniper density fell below 20%. We identified 84 791 ha within the study area occupied by juniper (3.6% of the total area), including 27 504 ha in deciduous forests (33% of deciduous forest area) and 38 738 ha in grasslands (6% of grassland area).

Mapping Urban Areas Using a Combination of Remote Sensing and Geolocation Data

Urban areas are essential to daily human life; however, the urbanization process also brings about problems, especially in China. Urban mapping at large scales relies heavily on remote sensing (RS) data, which cannot capture socioeconomic features well. Geolocation datasets contain patterns of human movement, which are closely related to the extent of urbanization. However, the integration of RS and geolocation data for urban mapping is performed mostly at the city level or finer scales due to the limitations of geolocation datasets. Tencent provides a large-scale location request density (LRD) dataset with a finer temporal resolution, and makes large-scale urban mapping possible. The objective of this study is to combine multi-source features from RS and geolocation datasets to extract information on urban areas at large scales, including night-time lights, vegetation cover, land surface temperature, population density, LRD, accessibility, and road networks. The random forest (RF) classifier is introduced to deal with these high-dimension features on a 0.01 degree grid. High spatial resolution land cover (LC) products and the normalized difference built-up index from Landsat are used to label all of the samples. The RF prediction results are evaluated using validation samples and compared with LC products for four typical cities. The results show that night-time lights and LRD features contributed the most to the urban prediction results. A total of 176,266 km2 of urban areas in China were extracted using the RF classifier, with an overall accuracy of 90.79% and a kappa coefficient of 0.790. Compared with existing LC products, our results are more consistent with the manually interpreted urban boundaries in the four selected cities. Our results reveal the potential of Tencent LRD data for the extraction of large-scale urban areas, and the reliability of the RF classifier based on a combination of RS and geolocation data.

A Cluster Graph Approach to Land Cover Classification Boosting

When it comes to land cover classification, the process of deriving the land classes is complex due to possible errors in algorithms, spatio-temporal heterogeneity of the Earth observation data, variation in availability and quality of reference data, or a combination of these. This article proposes a probabilistic graphical model approach, in the form of a cluster graph, to boost geospatial classifications and produce a more accurate and robust classification and uncertainty product. Cluster graphs can be characterized as a means of reasoning about geospatial data such as land cover classifications by considering the effects of spatial distribution, and inter-class dependencies in a computationally efficient manner. To assess the capabilities of our proposed cluster graph boosting approach, we apply it to the field of land cover classification. We make use of existing land cover products (GlobeLand30, CORINE Land Cover) along with data from Volunteered Geographic Information (VGI), namely OpenStreetMap (OSM), to generate a boosted land cover classification and the respective uncertainty estimates. Our approach combines qualitative and quantitative components through the application of our probabilistic graphical model and subjective expert judgments. Evaluating our approach on a test region in Garmisch-Partenkirchen, Germany, our approach was able to boost the overall land cover classification accuracy by 1.4% when compared to an independent reference land cover dataset. Our approach was shown to be robust and was able to produce a diverse, feasible and spatially consistent land cover classification in areas of incomplete and conflicting evidence. On an independent validation scene, we demonstrated that our cluster graph boosting approach was generalizable even when initialized with poor prior assumptions.

Generating a series of land covers by assimilating the existing land cover maps

Land cover (LC), which describes the physical material of land surface, is an important parameter in earth system science. With more and more LC data available, this research aims to establish a data assimilation framework for integrating continuous LC time series based on the existing LC products. The framework is designed by borrowing the basic concept from data assimilation to find out the optimal LC time series based on existing observations. According to the observing system simulation experiments, the assimilation framework showed a good performance against noises and uncertainties. Moreover, after validation with the existing LC reference data, the averaged accuracy of the LC data produced using the assimilation framework was 73.7%, which is a great improvement compared to the existing LC maps considering the temporal continuity. This LC map assimilation framework would be useful for assessing the global LC changes and related researches.

SCaMF–RM: A Fused High-Resolution Land Cover Product of the Rocky Mountains

Land cover (LC) products, derived primarily from satellite spectral imagery, are essential inputs for environmental studies because LC is a critical driver of processes involved in hydrology, ecology, and climatology, among others. However, existing LC products each have different temporal and spatial resolutions and different LC classes that rarely provide the detail required by these studies. Using multiple existing LC products, we implement our Spatiotemporal Categorical Map Fusion (SCaMF) methodology over a large region of the Rocky Mountains (RM), encompassing sections of six states, to create a new LC product, SCaMF–RM. To do this, we must adapt SCaMF to address the prediction of LC in large space–time regions that present nonstationarities, and we add more flexibility in the LC classifications of the predicted product. SCaMF–RM is produced at two high spatial resolutions, 30 and 50 m, and a yearly frequency for the 30-year period 1983–2012. When multiple products are available in time, we illustrate how SCaMF–RM captures relevant information from the different LC products and improves upon flaws observed in other products. Future work needed includes an exhaustive validation not only of SCaMF–RM but also of all input LC products.