Search Results (1,989)

Analyses of regional carbon stock dynamics, particularly of spatial and temporal dynamics and their relationship with land use transitions, play a key role in the management of terrestrial ecosystem functions and the optimization of land resource allocation. This study focuses on Shandong Province, an important ecological security barrier along the eastern coast of China, to explore carbon stock changes and how land use modifications contributed to the chrono-spatial distribution of carbon stocks from 1990 to 2020, with additional forecasts up to 2040. Based on Natural Variation Conditions, Ecological Variation Conditions, and the City’s Variation Conditions, the results indicate a downward trend in carbon stocks across Shandong Province, from 2661.87 × 10⁶ t in 1990 to 2380.02 × 10⁶ t in 2020. Carbon stocks exhibit a highly uneven spatial distribution, with concentrations being notably higher in the central and eastern regions. Cities are classified based on their carbon stock level: high carbon stock cities (Linyi, Weifang, Yantai), large carbon stock cities (Jinan, Jining, Qingdao, Dezhou, Binzhou, Liaocheng, Taian, Zibo, Dongying), and cities with general carbon stock levels (Weihai, Rizhao, Zaozhuang). The major driver of carbon stock decline is the conversion of ecological lands into urban areas, with cultivated lands and forests being the primary carbon storage contributors. Projections suggest that under the City’s Variation Conditions, carbon stocks will decrease from 2380.02 × 10⁶ t in 2020 to 1654.16 × 10⁶ t by 2040, while Carbon stocks will rise from 2380.02 × 10⁶ t to 2430.56 × 10⁶ t under the Ecological Variation Conditions. A significant disparity in carbon sink potential is found across cities, which are divided into high carbon sink potential cities (Yantai, Dezhou, Weifang, Qingdao, Jinan), large carbon sink potential cities (Binzhou, Weihai, Zibo, Liaocheng, Dongying, Linyi, Taian, Rizhao, Zaozhuang), and general potential cities (Jining, Heze). The insights gained from this study are essential for promoting the conservation of regional terrestrial ecosystems, directing land use policy development, and supporting sustainable development initiatives in Shandong Province. Full article

Household energy consumption is one of the major drivers of carbon emissions, and an in-depth analysis of its influencing factors, along with forecasting carbon emission trajectories, is crucial for achieving China’s carbon emission targets. This study reviews the research progress on urban household energy-related carbon emissions (HErC) in China since 2000, with a focus on the latest developments in influencing factors. The study categorizes these factors into five major groups: household characteristics, economic attributes, energy consumption features, awareness and norms, and policies and interventions. The findings indicate that income levels, energy efficiency, and household size are the key determinants of urban HErC of China and are commonly used as core assumptions in scenario-based forecasts of emission trends. In addition, although environmental awareness and government services have increasingly garnered attention, their specific effects require further investigation due to the challenges in quantification. A synthesis of existing forecasting studies suggests that, without the implementation of effective measures, HErC will continue to rise, and the peak emission period will be delayed. Enhancing building and energy efficiency, promoting low-carbon consumption and clean energy applications, and implementing multidimensional coordinated policies are considered the most effective pathways for emission reduction. Full article

Global ecosystems, particularly in biodiversity-rich tropical rainforests, are increasingly under pressure from human activities. As socio-economic development continues and populations steadily grow, the effective planning of areas surrounding national parks has become a global challenge. This study, based on remote sensing data and utilizing landscape ecology tools, such as ArcGIS 10.8, GeoDa 1.20, and Fragstats 4.2, combines spatial statistical methods, trend analysis, and the Hurst index to conduct a long-term analysis and forecast future trends in vegetation ecological quality indicators (VEQI) and landscape pattern changes within and around the Hainan Tropical Rainforest National Park. VEQI changes across various buffer zones were also assessed. Our results show that both arable and built-up land increased, especially from 2002 to 2022. Arable land decreased from 5566.8 km² to 4796.8 km², then increased to 5904.6 km²; built-up land expanded from 163.97 km² to 314.59 km², reflecting urbanization. Spatiotemporal analysis revealed that 42.54% of the study area experienced significant VEQI changes, with a 24.05% increase (mainly in the northwest) and an 18.49% decrease (mainly in the southeast). The VEQI improvements were consistent across all buffer zones, with the most significant growth in the 7.5 km zone. Landscape indices indicated high fragmentation in coastal areas, while inland areas remained stable, reflecting the tension between conservation and urbanization. These findings provide a theoretical basis for future ecological development and buffer zone policies in the park. Full article

Wave power (WP) is a strategic oceanic resource. Previous studies have extensively researched the long-term variations in WP in the South China Sea (SCS) for energy planning and utilization. This study extends the analysis of long-term trends to the last year based on ERA5 (European Centre for Medium-Range Weather Forecasts Reanalysis v5) reanalysis data from 1979 to 2024. Our results mainly indicate that the trends in WP after 2011 are significantly different from those before 2011. Before 2011, the WP in the SCS primarily showed an increasing trend, but, after 2011, it shifted to a decreasing trend. This trend has seasonal differences, manifested as being consistent with the annual trend in winter and spring while being inconsistent with the annual trend in summer and autumn. It indicates that the opposite trend in WP before and after 2011 was mainly the result of WP variations in winter and spring. To illustrate the driving factor for the WP’s variations, the contemporary long-term trend of the wind fields is systematically analyzed. Analysis results reveal that, regardless of seasonal differences or spatial distribution, the two trends are consistent in most situations, indicating that wind fields are the dominant factor for the long-term variations in WP. Meanwhile, the effects of the wind fields on the WP variations can also be modulated by environmental factors such as oceanic swell propagation and local topography. This study contributes to the knowledge of the latest trends and driving factors regarding the WP in the SCS. Full article

Compared to natural and shale gas, studies on predicting production specific to coalbed methane (CBM) are still relatively limited, and mainly use decline curve methods such as Arps, Stretched Exponential Decline Model, and Duong’s model. In recent years, machine learning (ML) methods applied to CBM production prediction have focused on the significant data characteristics of production, achieving more accurate predictions. However, throughout the application process, these models require a large amount of data for training and can only achieve accurate forecasts over a short period, such as 30 days. This study constructs a hybrid ML model by integrating a long short-term memory (LSTM) network and Transformer architecture. The model is trained using the mean absolute error (MAE) loss function, optimized using the Adam optimizer, and finally evaluated using metrics such as MAE, root mean square error (RMSE), and R squared (R²) scores. The results show that the LSTM-Attention (LSTM-A) hybrid model based on small training datasets can accurately capture the CBM production trend and is superior to traditional methods and the LSTM model regarding prediction accuracy and effective prediction time interval. The methodologies established and the results obtained in this study are of great significance to accurately predict CBM production. It is also helpful to better understand the mechanisms of CBM production. Full article

Global trends in health, climate, and population growth drive the demand for high-nutrient plants like spinach, which thrive under controlled conditions with minimal resources. Despite technological advances in agriculture, current systems often rely on traditional methods and need robust computational models for precise plant growth forecasting. Optimizing vegetable growth using advanced agricultural and computational techniques, addressing challenges in food security, and obtaining efficient resource utilization within urban agriculture systems are open problems for humanity. Considering the above, this paper presents an enclosed agriculture system for growth and modeling spinach of the Viroflay (Spinacia oleracea L.) species. It encompasses a methodology combining data science, machine learning, and mathematical modeling. The growth system was built using LED lighting, automated irrigation, temperature control with fans, and sensors to monitor environmental variables. Data were collected over 60 days, recording temperature, humidity, substrate moisture, and light spectra information. The experimental results demonstrate the effectiveness of polynomial regression models in predicting spinach growth patterns. The best-fitting polynomial models for leaf length achieved a minimum Mean Squared Error (MSE) of 0.158, while the highest MSE observed was 1.2153, highlighting variability across different leaf pairs. Leaf width models exhibited improved predictability, with MSE values ranging from 0.0741 to 0.822. Similarly, leaf stem length models showed high accuracy, with the lowest MSE recorded at 0.0312 and the highest at 0.3907. Full article

This study aims to investigate how innovation development drives green finance in the Visegrad countries by analyzing the role of R&D investments, high-tech trade, and patent activity in attracting greenfield investments. Using a vector autoregression (VAR) model with data from 2007 to 2022, this study employs forecasting techniques, impulse response functions, and variance decomposition analyses to assess the dynamic relationship between innovation and green financial flows. The findings reveal that R&D expenditures are the strongest driver of green investments, explaining over 93% of the variance in Poland and Hungary. High-tech trade significantly influences investment trends, contributing up to 84% of the variance in the Czech Republic, while patent applications initially boost greenfield investments but show diminishing returns over time. Although innovation-driven investments remain stable overall, the impact of trade and patents varies across countries, reflecting regional differences. This study identifies key challenges, such as commercialization gaps and policy disparities, highlighting the need for targeted financial and innovation policies. To sustain green finance growth, policymakers should expand R&D funding, strengthen trade infrastructure, and enhance intellectual property commercialization. Additionally, financial institutions and investors should play a more active role in developing green investment markets to support long-term economic resilience and sustainability. Full article

In the wave of digital transformation and Industry 4.0, accurate time series forecasting has become critical across industries such as manufacturing, energy, and finance. However, while deep learning models offer high predictive accuracy, their lack of interpretability often undermines decision-makers’ trust. This study proposes a linear time series model architecture based on seasonal decomposition. The model effectively captures trends and seasonality using an additive decomposition, chosen based on initial data visualization, indicating stable seasonal variations. An augmented feature generator is introduced to enhance predictive performance by generating features such as differences, rolling statistics, and moving averages. Furthermore, we propose a gradient-based feature importance method to improve interpretability and implement a gradient feature elimination algorithm to reduce noise and enhance model accuracy. The approach is validated on multiple datasets, including order demand, energy load, and solar radiation, demonstrating its applicability to diverse time series forecasting tasks. Full article

The Landsat archive stands as one of the most critical datasets for studying landscape change, offering over 50 years of imagery. This invaluable historical record facilitates the monitoring of land cover and land use changes, helping to detect trends in and the dynamics of the Earth’s system. However, the relatively low temporal frequency and irregular clear-sky observations of Landsat data pose significant challenges for multi-temporal analysis. To address these challenges, this research explores the application of a closed-form continuous-depth neural network (CFC) integrated within a recurrent neural network (RNN) called CFC-mmRNN for reconstructing historical Landsat time series in the Canadian Prairies region from 1985 to present. The CFC method was evaluated against the continuous change detection (CCD) method, widely used for Landsat time series reconstruction and change detection. The findings indicate that the CFC method significantly outperforms CCD across all spectral bands, achieving higher accuracy with improvements ranging from 33% to 42% and providing more accurate dense time series reconstructions. The CFC approach excels in handling the irregular and sparse time series characteristic of Landsat data, offering improvements in capturing complex temporal patterns. This study underscores the potential of leveraging advanced deep learning techniques like CFC to enhance the quality of reconstructed satellite imagery, thus supporting a wide range of remote sensing (RS) applications. Furthermore, this work opens up avenues for further optimization and application of CFC in higher-density time series datasets such as MODIS and Sentinel-2, paving the way for improved environmental monitoring and forecasting. Full article

The power-generation mix of future grids will be quite diversified with the ever-increasing share of renewable energy technologies. Therefore, the prediction of electricity demand will become crucial for resource optimization and grid stability. Machine learning- and artificial intelligence-based methods are widely studied by researchers to tackle the demand forecasting problem. However, since the COVID-19 pandemic broke out, new challenges have surfaced for forecasting research. In such a short amount of time, significant shifts have emerged in electricity demand trends, making it apparent that the pandemic and the possibility of similar crises in the future have escalated the complexity of energy management problems. Motivated by the circumstances, this research presents an hour-ahead and day-ahead electricity demand forecasting benchmark using Deep Polynomial Neural Networks (DNN) and Gene Expression Programming (GEP) methods. The DNN and GEP algorithms utilize on-site electricity consumption data collected from a university hospital for over two years with a temporal granularity of 15-minute intervals. Quarter-hourly meteorological, calendar, and daily COVID-19 data, including new cases and cumulative cases divided by four restriction levels, were also considered. These datasets are used not only to predict the electricity demand but also to investigate the impact of the COVID-19 pandemic on the electricity consumption of the hospital. The hour-ahead and day-ahead nRMSE results show that the DNN outperforms the GEP by $8.27\%$ and $14.32\%$, respectively. For the computational times, the DNN appears to be much faster than the GEP by $82.83\%$ and $78.56\%$ in the hour-ahead and day-ahead forecasting, respectively. Full article

The advent of Industry 4.0 and the integration of Artificial Intelligence (AI) is transforming supply chain management (SCM), improving efficiency, resilience and strategic decision-making capabilities. This research study provides a comprehensive overview of AI applications in key SCM processes, including customer relationship management, inventory management, transportation networks, procurement, demand forecasting and risk management. AI technologies such as Machine Learning, Natural Language Processing and Generative AI offer transformative solutions to streamline logistics, reduce operational risk and improve demand forecasting. In addition, this study identifies barriers to AI adoption, such as implementation challenges, organizational readiness and ethical concerns, and highlights the critical role of AI in promoting supply chain visibility and resilience in the midst of global crises. Future trends emphasize human-centric AI, increasing digital maturity, and addressing ethical and security concerns. This review concludes by confirming the critical role of AI in shaping sustainable, flexible and resilient supply chains while providing a roadmap for future research and application in SCM. Full article

Wheat farming is of utter importance in every country around the world, since it is a critical crop that contributes to food security and rural development. Given this importance, this research work investigates trends, determinants, and spatial variability in Polish wheat production between the years 2004 and 2023, with a key interest in productivity, resilience, and sustainable farming. This work will apply spatiotemporal analysis, statistical modeling, and ARIMA forecasting in the identification of the environmental, economic, and policy factors that interact with the wheat yield. Determinants include climatic variables, temperature, and precipitation, production costs, and market prices. These results point to great regional disparities in yield; the apparently better regions, such as Opole and Pomerania, enjoy exceptionally good environmental conditions with good access to modern technology, while regions like Podlasie and Subcarpathia are characterized by poor soil quality and a shortage of resources. This paper has pointed out the need for differential intervention policies that could contribute to reducing such yield gaps, increasing resilience to climate change, and hence contributing to sustainable wheat production growth. The stabilized yields projected underline the resilience of the sector to various challenges, from climate variability to market fluctuations. The results also indicate how innovative practices, supported by enabling policy frameworks, are essential in the promotion of wheat production in an environmentally friendly way, such as precision agriculture. The present research work will provide useful tools for policymakers, researchers, and other stakeholders by providing active insights into how to achieve equitable and sustainable agricultural development in Poland. Full article

Long-term runoff projection and uncertainty estimates can provide both the changing trends and confidence intervals of water resources, provide basic information for decision makers, and reduce risks for water resource management. In this paper, a grid-based runoff projection and uncertainty framework was proposed through input selection and long short-term memory (LSTM) modelling coupled with uncertainty analysis. We simultaneously considered dynamic variables and static variables in the candidate input combinations. Different input combinations were compared. We employed LSTM to develop a relationship between monthly runoff and the selected variables and demonstrated the improvement in forecast accuracy through comparison with the MLR, RBFNN, and RNN models. The LSTM model achieved the highest mean Kling–Gupta Efficiency (KGE) score of 0.80, representing respective improvements of 45.45%, 33.33%, and 2.56% over the other three models. The uncertainty sources originating from the parameters of the LSTM models were considered, and the Monte Carlo approach was used to provide uncertainty estimates. The framework was applied to the Yellow River Source Area (YRSR) at the 0.25° grid scale to better show the temporal and spatial features. The results showed that extra information about static variables can improve the accuracy of runoff projections. Annual runoff tended to increase, with projection ranges of 148.44–296.16 mm under the 95% confidence level, under various climate scenarios. Full article

Over the last 25 years, research on geoparks has moved from basic research to comprehensive multidisciplinary studies related to the creation and development of geoparks, integrating the principle of sustainability. This research focuses on exploring geoparks as the core subject. The aim of this study is to synthesize the heterogeneous body of knowledge about geoparks in an exhaustive way by leveraging a multi-database bibliometric approach. The methodology applied is based on quantitative bibliometric analysis using R, including its application for non-coders and ensuring reliability with the PRISMA Statement framework. Ten databases were taken as the sources of research papers: Web of Science, Scopus, PubMed, Nature Journals, SpringerLink, Taylor & Francis, Wiley Journals, IEEE Xplore, and CABI. The method we used has limitations, providing a restricted number of trends aligned and scaled to the database boundary conditions used in analysis. The main goals of quantitative bibliometric analysis are as follows: (1) The impact of data integration—Evaluating how merging the data from the ten databases improves research coverage. (2) Global research trends—Identifying the evolution of geopark-related studies over time. (3) Three-year forecast—Predicting the upcoming research directions using a polynomial regression model. (4) Academic performance—Assessing geographical distribution, citation impact, and productivity using bibliometric laws. (5) Conceptual contribution—Identifying the key research themes that drive future studies and potential areas for exploration. Among these, we highlighted the key elements. The integration of the ten databases provides 63% greater insight into scientific research compared to that of the Web of Science (WoS) database. Geographically, the scientific output spans 102 countries, with China leading in production over the last two decades. The most impactful paper has accumulated 768 citations, while Ruben D.A. and Wu Fandong emerge as the most prolific authors. According to the bibliometric law, the core source of scientific output is Geoheritage. The future research directions are expected to address global challenges, particularly natural disasters in alignment with the Sustainable Development Goals (SDGs). Additionally, GIS-based subtopics leveraging advanced technologies for analyzing, mapping, and promoting geological resources represent a promising area for further exploration. The projections indicate that by the end of 2026, scientific production in this field could reach 5226 published papers, underscoring the growing significance of geopark research and interdisciplinary advancements. Full article

This paper introduces a novel hybrid forecasting model for tourism demand that combines Bidirectional Long Short-Term Memory (BiLSTM) and Transformer networks, addressing the challenge of capturing both short-term fluctuations and long-term trends in complex tourism data. Unlike traditional models, such as ARIMA, which often struggle with nonlinear patterns, our hybrid approach leverages the sequential learning capabilities of BiLSTM and the self-attention mechanism of the Transformer to effectively model intricate temporal dependencies. Our experiments on Thailand’s domestic tourism data showed that the hybrid model outperformed traditional methods and standalone deep learning models, where it achieved a 12% reduction in the RMSE, a 15% reduction in the MAE, and a 10% increase in the $R^2$. This improved accuracy offers significant practical benefits for sustainable tourism, enabling policymakers and tourism managers to optimize resource allocation, anticipate peak season demand, and develop strategies to mitigate over-tourism. The model’s robustness and adaptability make it a valuable tool for data-driven decision-making in the tourism sector. Full article