Retrieval of subsurface dissolved oxygen from surface oceanic parameters based on machine learning.

Oceanic dissolved oxygen (DO) is crucial for oceanic material cycles and marine biological activities. However, obtaining subsurface DO values directly from satellite observations is limited due to the restricted observed depth. Therefore, it is essential to develop a connection between surface oceanic parameters and subsurface DO values. Machine learning (ML) methods can effectively grasp the complex relationship between input attributes and target variables, making them a valuable approach for estimating subsurface DO values based on surface oceanic parameters. In this study, the potential of ML methods for subsurface DO retrieval is analyzed. Among the selected ML methods, namely support vector regression (SVR), random forest (RF) regression, and extreme gradient boosting (XGBoosting) regression, the RF method generally demonstrates superior performance. As the depth increases, the accuracy of DO estimates tends to initially decrease, then gradually improve, with the poorest performance occurring at the depth of 600 dbar. The range of determination coefficients (R2) and root mean square error (RMSE) values based on the test dataset at different depths lies between 0.53 and 47.59 µmol/kg to 0.99 and 4.01 µmol/kg. In addition, compared to sea surface salinity (SSS) and sea surface chlorophyll-a (SCHL), sea surface temperature (SST) plays a more significant role in DO retrieval. Finally, compared to the pelagic interactions scheme for carbon and ecosystem studies (PISCES) model, the RF method achieves higher retrieval accuracies at depths above 700 dbar. In the deep ocean, the primary differences in DO values obtained from the RF method and the PISCES model-based method are noticeable in the vicinity of the equatorial region.

Estimation of the Ecosystem Service Value of the Yellow River Delta-Laizhou Bay Coastal Zone Considering Regional Differences and Social Development.

With economic and societal development, the ecological environment of the Yellow River Delta-Laizhou Bay coastal zone has been seriously damaged. Exploring the changes in land use and ecosystem service value (ESV) is essential to ecological construction of the region. The random forest classification method was used for land cover interpretation of the four periods of remote sensing images in the study area from 1990 to 2020. Newly calculated regional difference coefficients and social development coefficients were used to construct a dynamic ESV assessment model and to study its changes from overall and sea‒land gradient perspectives. The results showed that construction land, salt pans, aquaculture ponds, and inland water masses expanded rapidly, while cropland, tidal flats, and shallow waters shrank sharply over the past 30 years. The ESV in the study area has continued to decrease from 34.47 billion yuan in 1990 to 25.23 billion yuan in 2020, a total decrease of 9.23 billion yuan. This is mostly due to the encroachment of construction land, salt pans, and aquaculture ponds, and the flow of ecosystem services from high-value land cover types (tidal flats, herbaceous wetlands, and cropland) to medium- and low-value land cover types. Moreover, the land cover transfer and ESVs exhibited a decreasing trend from sea to land, with significant sea-land gradient differences. Land conversion is most common in the 0-15 km coastal zone, mainly from natural wetlands to artificial wetlands, where the ESV also decreases rapidly. Considering the regional differences and social development in this paper, the ESV of small-scale areas can be reasonably evaluated to explore the characteristics and causes of changes in land use and ESVs, which can provide an important reference for ecological protection and land use management in the region.

Optimized spatial and temporal pattern for coral bleaching heat stress alerts for China's coral reefs.

Most studies on coral bleaching alerts use common Degree Heating Week (DHW) thresholds; however, these may underestimate historical patterns of heat stress for coral reef ecosystems. Taking an optimized DHW threshold for coral bleaching alerts for Coral Reef Watch (CRW) and Coral Reef Temperature Anomaly Database (CoRTAD) products, we analyzed the precise spatial and temporal pattern of heat stress on China's coral reefs from 2010 to 2021 in the South China Sea (SCS) and the Beibu Gulf (BG). We compared acute heat stress using common and optimized thresholds. Results indicated that the ocean warming rate in 2010-2021 was approximately 0.43 ± 0.22 °C/10a, showing a significant increase in the northern SCS and the BG. More severe bleaching events were predicted by the optimized thresholds and the high-frequency areas were mainly in the northern SCS. The number and intensity of years with severe heat stress anomalies was in the order 2020 > 2014 > 2010 > 2015. Heat stress duration was the longest in the Xisha Islands among offshore archipelagos, and longest in 2020-2021 in Weizhou Island in BG in the relative high-latitude inshore reefs. These abnormal events were mainly caused by El Niño, but La Niña was also involved in 2020.

Prediction of fishing intensity and trends across South China Sea biogeographic zones.

The volume of industrial fishing in the South China Sea ranks among the top global sustainable fisheries concerns of the Food and Agriculture Organization (FAO). To better understand the scale of management challenges, biogeographic zones of the SCS were characterized, and within each a multivariate GAM (General Additive Model) was fitted to predict and map the complete fishing activities from 2017 to 2020. Model variables, some incomplete or with gaps, included: VIIRS DNB night-time light imagery; Global Fisheries Watch (GFW) data; satellite Ocean Colour; Sea Surface Temperature; and bathymetry data. Four biogeographic zones with differing fishing patterns and trends were identified. We used cross-validation and the GAM model's own tuning method for model prediction accuracy determination, which performed well in four biogeographic zones (R2 respectively: 0.62, 0.68, 0.74 and 0.71). High-intensity fishing grounds are mainly distributed in offshore continental shelf areas. From 2017 to 2019, high-intensity fishing grounds were located near the Beibu Gulf of Vietnam, south Vietnam, part of the Gulf of Thailand and the central Java Sea, where fishing effort greater than 50 h exceeded average annual SCS fishing intensity for several years. By season, intensity and extent of fishing in Spring were largest. In 2020, due to the impact of COVID-19, except for Spring, fishing volume generally decreased. Our experimental results provide new insights and an adaptable biogeographic modelling methodology to map the scale and intensity of regional fishing activities more accurately and completely. This more comprehensive database, that takes account of intrinsic biogeographic fishery context, will help improve and strengthen the regulation of fishing activities around the world.

Widespread global disparities between modelled and observed mid-depth ocean currents.

The mid-depth ocean circulation is critically linked to actual changes in the long-term global climate system. However, in the past few decades, predictions based on ocean circulation models highlight the lack of data, knowledge, and long-term implications in climate change assessment. Here, using 842,421 observations produced by Argo floats from 2001-2020, and Lagrangian simulations, we show that only 3.8% of the mid-depth oceans, including part of the equatorial Pacific Ocean and the Antarctic Circumpolar Current, can be regarded as accurately modelled, while other regions exhibit significant underestimations in mean current velocity. Knowledge of ocean circulation is generally more complete in the low-latitude oceans but is especially poor in high latitude regions. Accordingly, we propose improvements in forecasting, model representation of stochasticity, and enhancement of observations of ocean currents. The study demonstrates that knowledge and model representations of global circulation are substantially compromised by inaccuracies of significant magnitude and direction, with important implications for modelled predictions of currents, temperature, carbon dioxide sequestration, and sea-level rise trends.

Global Container Port Network Linkages and Topology in 2021.

The maritime transport of containers between ports accounts for the bulk of global trade by weight and value. Transport impedance among ports through transit times and port infrastructures can, however, impact accessibility, trade performance, and the attractiveness of ports. Assessments of the transit routes between ports based on performance and attractiveness criteria can provide a topological liner shipping network that quantifies the performance profile of ports. Here, we constructed a directed global liner shipping network (GLSN) of the top six liner shipping companies between the ports of Africa, Asia, North/South America, Europe, and Oceania. Network linkages and community groupings were quantified through a container port accessibility evaluation model, which quantified the performance of the port using betweenness centrality, the transport impedance among ports with the transit time, and the performance of ports using the Port Liner Shipping Connectivity Index. The in-degree and out-degree of the GLSN conformed to the power-law distribution, respectively, and their R-square fitting accuracy was greater than 0.96. The community partition illustrated an obvious consistence with the actual trading flow. The accessibility evaluation result showed that the ports in Asia and Europe had a higher accessibility than those of other regions. Most of the top 30 ports with the highest accessibility are Asian (17) and European (10) ports. Singapore, Port Klang, and Rotterdam have the highest accessibility. Our research may be helpful for further studies such as species invasion and the planning of ports.

Performance of air temperature from ERA5-Land reanalysis in coastal urban agglomeration of Southeast China.

Near-surface air temperature is an important indicator of climate change and extreme events. ERA5-Land reanalysis products feature finer spatial and temporal resolutions, and have been widely adopted in global climate-related research. However, the performance of ERA5-Land air temperature data in coastal urban agglomerations has received little attention. In this study, a comprehensive evaluation is conducted in the Guangdong-Hong Kong-Macau Greater Bay Area (GBA) using the observations of 1080 automatic weather stations in 2018 as reference. Generally, ERA5-Land underestimates temperature (an average bias of 0.90 °C), and performs better at low temperatures than at high temperatures. At the station level, it is observed that the correlation shows a strong positive linear relationship with the distance to the coastline in summer, and that the bias increases with increasing altitude throughout the year. With respect to different land cover types, data accuracy over urban and built-up lands is the lowest. The spatial pattern of ERA5-Land is generally consistent with that of stations but relatively poor in urban areas. In addition, ERA5-Land properly captures daily and monthly variations, as well as intraday temperature fluctuations. These conclusions provide a reference for the implementation of ERA5-Land in coastal urban agglomerations.

Reclamation and Ecological Service Value Evaluation of Coastal Wetlands Using Multispectral Satellite Imagery.

There are special locational value and natural resources in coastal wetlands. Studying their changes and evaluating their ecosystem service value (ESV) is beneficial for protecting the ecology of coastal wetlands and for maintaining sustainable human development. In this paper, the coastal wetland of Jiaozhou Bay is selected as the research area, an object-oriented method is used to extract shoreline and wetland information, and the coastal wetland reclamation process in Jiaozhou Bay is evaluated. The value equivalent method and market value method are used to evaluate the service value of wetland ecosystems from the perspective of ecological economics. The results show that the reclamation area of Jiaozhou Bay reached 75.2 km2 in 40 years, with nearly 23% of the bay area eroding. Reclamation engineering, estuary engineering, policy implementation and urbanization are the main factors affecting the changes in the Jiaozhou Bay wetland, and the main direction of wetland succession is natural wetlands→artificial wetlands→nonwetlands. Wetland reclamation in Jiaozhou Bay has led to the continuous extension of the coastline to the sea, especially during the 2005-2020 period, and the wetland area has declined in area by 116 km2. The changes in the wetland in the past 40 years have affected the changes in the ESV of Jiaozhou Bay, and there have been different synergistic/trade-off relationships in different periods. This research provides data to support the comprehensive ecological management of coastal areas, which is conducive to maximizing the utilization value of wetlands and promoting wetland protection.

Rapid greening response of China's 2020 spring vegetation to COVID-19 restrictions: Implications for climate change.

The 2019 novel coronavirus pandemic (COVID-19) negatively affected global public health and socioeconomic development. Lockdowns and travel restrictions to contain COVID-19 resulted in reduced human activity and decreased anthropogenic emissions. However, the secondary effects of these restrictions on the biophysical environment are uncertain. Using remotely sensed big data, we investigated how lockdowns and traffic restrictions affected China's spring vegetation in 2020. Our analyses show that travel decreased by 58% in the first 18 days following implementation of the restrictions across China. Subsequently, atmospheric optical clarity increased and radiation levels on the vegetation canopy were augmented. Furthermore, the spring of 2020 arrived 8.4 days earlier and vegetation 17.45% greener compared to 2015-2019. Reduced human activity resulting from COVID-19 restrictions contributed to a brighter, earlier, and greener 2020 spring season in China. This study shows that short-term changes in human activity can have a relatively rapid ecological impact at the regional scale.

Development of a comprehensive assessment model for coral reef island carrying capacity(CORE-CC).

Coral reef islands provide precious living space and valuable ecological services for human beings, and its sustainability cannot be ignored under the pressure of human activities. Carrying capacity (CC) assessment has gradually become an important means to measure sustainability of islands. However, there is little comprehensive evaluation of the carrying capacity of coral reef islands, and traditional evaluation methods are difficult to express the social-ecological characteristics of coral reef islands. The present paper proposes a comprehensive assessment model for coral reef island carrying capacity (CORE-CC) which comprises dimensions of resources supply, environmental assimilative, ecosystem services, and socio-economic supporting. According to the characteristics of the coral reef islands, the core factors and indicators of each dimension are selected and the corresponding assessment index system of "pressure-support" is constructed. The assessment involves (1) identification of carrying dimensions and core factors, (2) pressure/support measurement and (3) assessment of carrying state. A case study is conducted in Zhaoshu Island of China, demonstrating the applicability of CORE-CC model and serving as a reference for adaptive management.

Mapping lead concentrations in urban topsoil using proximal and remote sensing data and hybrid statistical approaches.

Due to rapid urbanization in China, lead (Pb) continues to accumulate in urban topsoil, resulting in soil degradation and increased public exposure. Mapping Pb concentrations in urban topsoil is therefore vital for the evaluation and control of this exposure risk. This study developed spatial models to map Pb concentrations in urban topsoil using proximal and remote sensing data. Proximal sensing reflectance spectra (350-2500 nm) of soils were pre-processed and used to calculate the principal components as landscape factors to represent the soil properties. Other landscape factors, including vegetation and land-use factors, were extracted from time-sequential Landsat images. Two hybrid statistical approaches, regression kriging (RK) and geographically weighted regression (GWR), were adopted to establish prediction models using the landscape factors. The results indicated that the use of landscape factors derived from combined remote and proximal sensing data improved the prediction of Pb concentrations compared with useing these data individually. GWR obtained better results than RK for predicting soil Pb concentration. Thus, joint proximal and remote sensing provides timely, easily accessible, and suitable data for extracting landscape factors.

Spatial-temporal changes of forests and agricultural lands in Malaysia from 1990 to 2017.

Forests and agricultural lands are the main resources on the earth's surface and important indicators of regional ecological environments. In this paper, Landsat images from 1990 and 2017 were used to extract information on forests in Malaysia based on a remote-sensing classification method. The spatial-temporal changes of forests and agricultural lands in Malaysia between 1990 and 2017 were analyzed. The results showed that the natural forests in Malaysia decreased by 441 Mha, a reduction of 21%. The natural forests were mainly converted into plantations in Peninsular Malaysia and plantations and secondary forests in East Malaysia. The area of agricultural lands in Malaysia increased by 55.7%, in which paddy fields increased by 1.1% and plantations increased by 98.2%. Paddy fields in Peninsular Malaysia are mainly distributed in the north-central coast and the Kelantan Delta. The agricultural land in East Malaysia is dominated by plantations, which are mainly distributed in coastal areas. The predictable areas of possible expansion for paddy fields in Peninsular Malaysia's Kelantan (45.2%) and Kedah (16.8%) areas in the future are large, and in addition, the plantations in Sarawak (44.7%) and Sabah (29.6%) of East Malaysia have large areas for expansion. The contradiction between agricultural development and protecting the ecological environment is increasingly prominent. The demand for agriculture is expected to increase further and result in greater pressures on tropical forests. Governments also need to encourage farmers to carry out existing land development, land recultivation, or cooperative development to improve agricultural efficiency and reduce the damage to natural forests.

Machine learning-based detection of soil salinity in an arid desert region, Northwest China: A comparison between Landsat-8 OLI and Sentinel-2 MSI.

Accurate assessment of soil salinization is considered as one of the most important steps in combating global climate change, especially in arid and semi-arid regions. Multi-spectral remote sensing (RS) data including Landsat series provides the potential for frequent surveys for soil salinization at various scales and resolutions. Additionally, the recently launched Sentinel-2 satellite constellation has temporal revisiting frequency of 5 days, which has been proven to be an ideal approach to assess soil salinity. Yet, studies on detailed comparison in soil salinity tracking between Landsat-8 OLI and Sentinel-2 MSI remain limited. For this purpose, we collected a total of 64 topsoil samples in an arid desert region, the Ebinur Lake Wetland National Nature Reserve (ELWNNR) to compare the monitoring accuracy between Landsat-8 OLI and Sentinel-2 MSI. In this study, the Cubist model was trained using RS-derived covariates (spectral bands, Tasseled Cap transformation-derived wetness (TCW), and satellite salinity indices) and laboratory measured electrical conductivity of 1:5 soil:water extract (EC). The results showed that the measured soil salinity had a significant correlation with surface soil moisture (Pearson's r = 0.75). The introduction of TCW generated satisfactory estimating performance. Compared with OLI dataset, the combination of MSI dataset and Cubist model yielded overall better model performance and accuracy measures (R2 = 0.912, RMSE = 6.462 dS m-1, NRMSE = 9.226%, RPD = 3.400 and RPIQ = 6.824, respectively). The differences between Landsat-8 OLI and Sentinel-2 MSI were distinguishable. In conclusion, MSI image with finer spatial resolution performed better than OLI. Combining RS data sets and their derived TCW within a Cubist framework yielded accurate regional salinity map. The increased temporal revisiting frequency and spectral resolution of MSI data are expected to be positive enhancements to the acquisition of high-quality soil salinity information of desert soils.

Automatic Extraction of Offshore Platforms in Single SAR Images Based on a Dual-Step-Modified Model.

The quantity and location of offshore platforms are of great significance for marine oil spill monitoring and offshore oil-gas development. In the past, multiphase medium- and low-resolution optical or radar images have been used to remove the interference of ship targets based on the static position of a platform to extract the offshore platform, resulting in large demands and high image data costs. According to the difference in shape between offshore platforms (not elongated) and ships (elongated shapes) in SAR (synthetic aperture radar) images, this paper proposes an automatic extraction method for offshore platforms in single SAR images based on a dual-step-modified model. First, the two-parameter CFAR (constant false alarm rate) algorithm was used to detect the possible offshore platform targets; then, the Hough transform was introduced to detect and eliminate ship targets with linear structures. Finally, the final offshore platform was obtained. Experiments were carried out in four study areas in the Beibu Gulf basin and the Pearl River estuary basin in the northern South China Sea. The results show that the method has a good extraction effect in the above research area, and the extraction accuracy rate of offshore platforms is 86.75%. A single SAR image can obtain satisfactory extraction results, which greatly saves on image data cost.

Offshore Platform Extraction Using RadarSat-2 SAR Imagery: A Two-Parameter CFAR Method Based on Maximum Entropy.

The ability to determine the number and location of offshore platforms is of great significance for offshore oil spill monitoring and offshore oil and gas development. Considering the problem that the detection threshold parameters of the two-parameter constant false alarm rate (CFAR) algorithm require manual and repeated adjustment of the during the extraction of offshore platform targets, this paper proposes a two-parameter CFAR target detection method based on maximum entropy based on information entropy theory. First, a series of threshold parameters are obtained using the two-parameter CFAR algorithm for target detection. Then, according to the maximum entropy principle, the optimal threshold is estimated to obtain the target detection results of the possible offshore platform. Finally, the neighborhood analysis method is used to eliminate false alarm targets such as ships, and the final target of the offshore platform is obtained. In this study, we conducted offshore platform extraction experiments and an accuracy evaluation using data from the Pearl River Estuary Basin of the South China Sea. The results show that the proposed method for platform extraction achieves an accuracy rate of 97.5% and obtains the ideal offshore platform distribution information. Thus, the proposed method can objectively obtain the optimal target detection threshold parameters, greatly reduce the influence of subjective parameter setting on the extraction results during the target detection process and effectively extract offshore platform targets.

An evolving assessment model for environmental carrying capacity: A case study of coral reef islands.

Environmental carrying capacity (ECC) provides an insight into measuring sustainability of the vulnerable coral reef islands. However, an integrated assessment of ECC on the social-ecological system of reef islands is rarely existence. And conventional approaches miss addressing the difference of social development, which would lead to a misinterpretation of sustainable development of reef island system. This study develops an evolving model of RI-ECC which incorporates five specific development phases, and the assessment involves (1) identification and measurement of carrying components, (2) supply/demand surplus analysis of indicators and (3) ECC states determination. A case study is conducted in Zhaoshu Island of China, indicating the efficiency of RI-ECC model and serving as a reference for adaptive management.

Tree-based approach for exploring marine spatial patterns with raster datasets.

From multiple raster datasets to spatial association patterns, the data-mining technique is divided into three subtasks, i.e., raster dataset pretreatment, mining algorithm design, and spatial pattern exploration from the mining results. Comparison with the former two subtasks reveals that the latter remains unresolved. Confronted with the interrelated marine environmental parameters, we propose a Tree-based Approach for eXploring Marine Spatial Patterns with multiple raster datasets called TAXMarSP, which includes two models. One is the Tree-based Cascading Organization Model (TCOM), and the other is the Spatial Neighborhood-based CAlculation Model (SNCAM). TCOM designs the "Spatial node→Pattern node" from top to bottom layers to store the table-formatted frequent patterns. Together with TCOM, SNCAM considers the spatial neighborhood contributions to calculate the pattern-matching degree between the specified marine parameters and the table-formatted frequent patterns and then explores the marine spatial patterns. Using the prevalent quantification Apriori algorithm and a real remote sensing dataset from January 1998 to December 2014, a successful application of TAXMarSP to marine spatial patterns in the Pacific Ocean is described, and the obtained marine spatial patterns present not only the well-known but also new patterns to Earth scientists.

An Improved DINEOF Algorithm for Filling Missing Values in Spatio-Temporal Sea Surface Temperature Data.

In this study, an improved Data INterpolating Empirical Orthogonal Functions (DINEOF) algorithm for determination of missing values in a spatio-temporal dataset is presented. Compared with the ordinary DINEOF algorithm, the iterative reconstruction procedure until convergence based on every fixed EOF to determine the optimal EOF mode is not necessary and the convergence criterion is only reached once in the improved DINEOF algorithm. Moreover, in the ordinary DINEOF algorithm, after optimal EOF mode determination, the initial matrix with missing data will be iteratively reconstructed based on the optimal EOF mode until the reconstruction is convergent. However, the optimal EOF mode may be not the best EOF for some reconstructed matrices generated in the intermediate steps. Hence, instead of using asingle EOF to fill in the missing data, in the improved algorithm, the optimal EOFs for reconstruction are variable (because the optimal EOFs are variable, the improved algorithm is called VE-DINEOF algorithm in this study). To validate the accuracy of the VE-DINEOF algorithm, a sea surface temperature (SST) data set is reconstructed by using the DINEOF, I-DINEOF (proposed in 2015) and VE-DINEOF algorithms. Four parameters (Pearson correlation coefficient, signal-to-noise ratio, root-mean-square error, and mean absolute difference) are used as a measure of reconstructed accuracy. Compared with the DINEOF and I-DINEOF algorithms, the VE-DINEOF algorithm can significantly enhance the accuracy of reconstruction and shorten the computational time.