Tidal wetlands are expected to respond dynamically to global environmental change, but the extent to which wetland losses have been offset by gains remains poorly understood. We developed a global ...analysis of satellite data to simultaneously monitor change in three highly interconnected intertidal ecosystem types-tidal flats, tidal marshes, and mangroves-from 1999 to 2019. Globally, 13,700 square kilometers of tidal wetlands have been lost, but these have been substantially offset by gains of 9700 km
, leading to a net change of -4000 km
over two decades. We found that 27% of these losses and gains were associated with direct human activities such as conversion to agriculture and restoration of lost wetlands. All other changes were attributed to indirect drivers, including the effects of coastal processes and climate change.
•The integrated satellite imaging and data transmission scheduling problem is quite complex.•An improved genetic algorithm is presented.•The effectiveness of several operators is validated.•Our ...approach can obtain high-quality solutions within a limited time period.•Our approach performs well for the large scale optimization instances.
The study of Integrated Satellite Imaging and Data Transmission Scheduling Problem (ISIDTSP) has assumed increasing importance due to the growing number of satellites observing large quantities of targets and seeking to transmit their images to the ground stations. This paper formulates the ISIDTSP as a mixed integer programming model and develops an improved genetic algorithm. With regard to the individual representation, a novel idea of encoding and decoding is adopted to match the specific request with the corresponding satellite-ground resources, and a conception of conflicting request set is proposed to limit the chromosome length, thereby reducing the algorithmic time complexity. For the trade-off between diversity and convergence, several effective operators are introduced, including the population initialization based on the way of uniform design, the multi-point greedy mutation and the adaptive selection. Evaluations with two test cases demonstrate the efficiency of the proposed algorithm and show its ability to obtain high-quality solutions within an acceptable time period for the large scale optimization instances.
Accurate and comprehensive measurements of economic well-being are fundamental inputs into both research and policy, but such measures are unavailable at a local level in many parts of the world. ...Here we train deep learning models to predict survey-based estimates of asset wealth across ~ 20,000 African villages from publicly-available multispectral satellite imagery. Models can explain 70% of the variation in ground-measured village wealth in countries where the model was not trained, outperforming previous benchmarks from high-resolution imagery, and comparison with independent wealth measurements from censuses suggests that errors in satellite estimates are comparable to errors in existing ground data. Satellite-based estimates can also explain up to 50% of the variation in district-aggregated changes in wealth over time, with daytime imagery particularly useful in this task. We demonstrate the utility of satellite-based estimates for research and policy, and demonstrate their scalability by creating a wealth map for Africa's most populous country.
Precision agriculture is considered to be a fundamental approach in pursuing a low-input, high-efficiency, and sustainable kind of agriculture when performing site-specific management practices. To ...achieve this objective, a reliable and updated description of the local status of crops is required. Remote sensing, and in particular satellite-based imagery, proved to be a valuable tool in crop mapping, monitoring, and diseases assessment. However, freely available satellite imagery with low or moderate resolutions showed some limits in specific agricultural applications, e.g., where crops are grown by rows. Indeed, in this framework, the satellite's output could be biased by intra-row covering, giving inaccurate information about crop status. This paper presents a novel satellite imagery refinement framework, based on a deep learning technique which exploits information properly derived from high resolution images acquired by unmanned aerial vehicle (UAV) airborne multispectral sensors. To train the convolutional neural network, only a single UAV-driven dataset is required, making the proposed approach simple and cost-effective. A vineyard in Serralunga d'Alba (Northern Italy) was chosen as a case study for validation purposes. Refined satellite-driven normalized difference vegetation index (NDVI) maps, acquired in four different periods during the vine growing season, were shown to better describe crop status with respect to raw datasets by correlation analysis and ANOVA. In addition, using a K-means based classifier, 3-class vineyard vigor maps were profitably derived from the NDVI maps, which are a valuable tool for growers.
We have monitored a newly erupted volcanic island in the Kingdom of Tonga, unofficially known as Hunga Tonga Hunga Ha'apai, by means of relatively frequent high spatial resolution (~50 cm) satellite ...observations. The new ~1.8 km 2(exp) island formed as a tuff cone over the course of a month‐long hydromagmatic eruption in early 2015 in the Tonga‐Kermadec volcanic arc. Such ash‐dominated eruptions usually produce fragile subaerial landscapes that wash away rapidly due to marine erosion, as occurred nearby in 2009. Our measured rates of erosion are ~0.00256 km 3(exp) / year from derived digital topographic models. Preliminary measurements of the topographic expression of the primary tuff cone over ~30 months suggest a lifetime of ~19 years (and potentially up to 42 years). The ability to measure details of a young island's landscape evolution using satellite remote sensing has not previously been possible at these spatial and temporal resolutions.
Satellite data and aerial photos have proved to be useful in efficient conservation and management of mangrove ecosystems. However, there have been only very few attempts to demonstrate the ability ...of drone images, and none so far to observe vegetation (species-level) mapping. The present study compares the utility of drone images (DJI-Phantom-2 with SJ4000 RGB and IR cameras, spatial resolution: 5cm) and satellite images (Pleiades-1B, spatial resolution: 50cm) for mangrove mapping-specifically in terms of image quality, efficiency and classification accuracy, at the Setiu Wetland in Malaysia. Both object- and pixel-based classification approaches were tested (QGIS v.2.12.3 with Orfeo Toolbox). The object-based classification (using a manual rule-set algorithm) of drone imagery with dominant land-cover features (i.e. water, land, Avicennia alba, Nypa fruticans, Rhizophora apiculata and Casuarina equisetifolia) provided the highest accuracy (overall accuracy (OA): 94.0±0.5% and specific producer accuracy (SPA): 97.0±9.3%) as compared to the Pleiades imagery (OA: 72.2±2.7% and SPA: 51.9±22.7%). In addition, the pixel-based classification (using a maximum likelihood algorithm) of drone imagery provided better accuracy (OA: 90.0±1.9% and SPA: 87.2±5.1%) compared to the Pleiades (OA: 82.8±3.5% and SPA: 80.4±14.3%). Nevertheless, the drone provided higher temporal resolution images, even on cloudy days, an exceptional benefit when working in a humid tropical climate. In terms of the user-costs, drone costs are much higher, but this becomes advantageous over satellite data for long-term monitoring of a small area. Due to the large data size of the drone imagery, its processing time was about ten times greater than that of the satellite image, and varied according to the various image processing techniques employed (in pixel-based classification, drone >50 hours, Pleiades <5 hours), constituting the main disadvantage of UAV remote sensing. However, the mangrove mapping based on the drone aerial photos provided unprecedented results for Setiu, and was proven to be a viable alternative to satellite-based monitoring/management of these ecosystems. The improvements of drone technology will help to make drone use even more competitive in the future.
Abstract
In view of the “4·21” Indonesian submarine “Nanggala” crash event, an analysis of the marine environmental factors that may lead to the crash is carried out. The results show that the ocean ...internal wave is one possible cause of the disappearance of Indonesian submarines. The event is most likely that the submarine was subject to the so called “sea cliff” when encountering large-amplitude ocean internal waves, and at the same time, the officers and soldiers on board failed to take reasonable evasive measures in time, resulting in the submarine losing control and falling below the maximum safe diving depth, making the submarine crack under the pressure of the huge sea, resulting in a submarine accident. This conclusion is obtained based on the analysis of integrated ocean internal wave satellite images and numerical simulations.
Flooding affects more people than any other environmental hazard and hinders sustainable development. Investing in flood adaptation strategies may reduce the loss of life and livelihood caused by ...floods. Where and how floods occur and who is exposed are changing as a result of rapid urbanization4, flood mitigation infrastructure and increasing settlements in floodplains6. Previous estimates of the global flood-exposed population have been limited by a lack of observational data, relying instead on models, which have high uncertainty. Here we use daily satellite imagery at 250-metre resolution to estimate flood extent and population exposure for 913 large flood events from 2000 to 2018. We determine a total inundation area of 2.23 million square kilometres, with 255–290 million people directly affected by floods. We estimate that the total population in locations with satellite-observed inundation grew by 58–86 million from 2000 to 2015. This represents an increase of 20 to 24 per cent in the proportion of the global population exposed to floods, ten times higher than previous estimates. Climate change projections for 2030 indicate that the proportion of the population exposed to floods will increase further. The high spatial and temporal resolution of the satellite observations will improve our understanding of where floods are changing and how best to adapt. The global flood database generated from these observations will help to improve vulnerability assessments, the accuracy of global and local flood models, the efficacy of adaptation interventions and our understanding of the interactions between landcover change, climate and floods.
•Spaceborne one-shot hyperspectral imagery for crop mapping in cloudy season.•The three-dimensional CNN model is found to be competent for identifying crops.•The in-depth features are analyzed and ...visualized using the t-SNE technique.•Results show one-shot hyperspectral images is superior to multi-temporal images in cloudy season.
Crop mapping is essential for agricultural management, economic development planning, and ecological conservation. Remote sensing with a large field of view provides us with a potential technique for large-scale crop mapping. However, most of the previous studies have focused on multi-temporal crop mapping, requiring multiple imaging over a period of time, which is impossible in the cloudy season due to the absence of clear atmospheric windows. Recently, with the progress of spaceborne hyperspectral imaging technology, wide-width hyperspectral satellite images, which provide abundant spectral and spatial structure information, have made the precision crop mapping of large areas possible. This paper focuses on deep learning based crop mapping using one-shot hyperspectral satellite imagery, where three convolutional neural network (CNN) models, i.e., 1D-CNN, 2D-CNN, and 3D-CNN models, are applied for end-to-end crop mapping. In addition, a manifold learning based visualization approach, i.e., t-distributed stochastic neighbor embedding (t-SNE), is introduced to illustrate the discriminative ability of the deep semantic features extracted by the different CNN models. To demonstrate the advantages of one-shot hyperspectral satellite images, an experiment was designed to compare the crop mapping performance of different remote sensing data sources, where both mono-temporal and multi-temporal multispectral images (MSIs) of the same research area were introduced for a systematic comparison. The classification accuracy when using hyperspectral satellite images was found to reach more than 94%, which was much better than that when using mono-temporal MSIs, and was comparable to the result when using multi-temporal MSIs. These findings will be important for the application of hyperspectral data when mapping large-area crop landscapes, and they confirm the potential of CNN models, particularly 3D-CNN models, for crop recognition.
This letter investigates fully convolutional networks (FCNs) for the detection of informal settlements in very high resolution (VHR) satellite images. Informal settlements or slums are proliferating ...in developing countries and their detection and classification provides vital information for decision making and planning urban upgrading processes. Distinguishing different urban structures in VHR images is challenging because of the abstract semantic definition of the classes as opposed to the separation of standard land-cover classes. This task requires extraction of texture and spatial features. To this aim, we introduce deep FCNs to perform pixel-wise image labeling by automatically learning a higher level representation of the data. Deep FCNs can learn a hierarchy of features associated to increasing levels of abstraction, from raw pixel values to edges and corners up to complex spatial patterns. We present a deep FCN using dilated convolutions of increasing spatial support. It is capable of learning informative features capturing long-range pixel dependencies while keeping a limited number of network parameters. Experiments carried out on a Quickbird image acquired over the city of Dar es Salaam, Tanzania, show that the proposed FCN outperforms state-of-the-art convolutional networks. Moreover, the computational cost of the proposed technique is significantly lower than standard patch-based architectures.