Search Results (141)

Most reinforcement learning (RL) methods for portfolio optimization remain limited to single markets and a single algorithmic paradigm, which restricts their adaptability to regime shifts and heterogeneous conditions. This paper introduces a generalized version of the Modular Portfolio Learning System (MPLS), extending beyond its initial PPO backbone to integrate four RL algorithms: Proximal Policy Optimization (PPO), Deep Q-Network (DQN), Deep Deterministic Policy Gradient (DDPG), and Soft Actor-Critic (SAC). Building on its modular design, MPLS leverages specialized components for sentiment analysis, volatility forecasting, and structural dependency modeling, whose signals are fused within an attention-based decision framework. Unlike prior approaches, MPLS is evaluated independently on three major equity indices (S&P 500, DAX 30, and FTSE 100) across diverse regimes including stable, crisis, recovery, and sideways phases. Experimental results show that MPLS consistently achieved higher Sharpe ratios—typically +40–70% over Minimum Variance Portfolio (MVP) and Risk Parity (RP)—while limiting drawdowns and Conditional Value-at-Risk (CVaR) during stress periods such as the COVID-19 crash. Turnover levels remained moderate, confirming cost-awareness. Ablation and variance analyses highlight the distinct contribution of each module and the robustness of the framework. Overall, MPLS represents a modular, resilient, and practically relevant framework for risk-aware portfolio optimization. Full article

Sectoral vulnerability to distinct crisis types in small, open, and geopolitically exposed markets—such as Jordan—remains insufficiently quantified, constraining targeted policy design and portfolio allocation. This study’s primary purpose is to establish a transparent, comparable metric of sector-level market resilience that reveals how crisis typology reorders vulnerabilities and shapes recovery speed. Applying this framework, we assess Jordan’s equity market across three archetypal episodes—the Global Financial Crisis, the Arab Spring, and COVID-19—to clarify how shock channels reconfigure sectoral risk. Using daily Amman Stock Exchange sector indices (2001–2025), we estimate $GARCH\left(\mathrm{1,1}\right)$ models for each sector–crisis window and summarize volatility dynamics by persistence $(\alpha +\beta )$, interpreted as an inverse proxy for resilience; complementary diagnostics include maximum drawdown and days-to-recovery, with nonparametric (Kruskal–Wallis) and rank-based (Spearman, Friedman) tests to evaluate within-crisis differences and cross-crisis reordering. Results show pronounced heterogeneity in every crisis and shifting sectoral rankings: financials—especially banking—display the highest persistence during the GFC; tourism and transportation dominate during COVID-19; and tourism/electric-related industries are most persistent around the Arab Spring. Meanwhile, food & beverages, pharmaceuticals/medical, and education recurrently exhibit lower persistence. Higher persistence aligns with slower post-shock normalization. We conclude that resilience is sector-specific and contingent on crisis characteristics, implying targeted policy and portfolio responses; regulators should prioritize liquidity backstops, timely disclosure, and contingency planning for fragile sectors, while investors can mitigate crisis risk via dynamic sector allocation and volatility-aware risk management in emerging markets. Full article

Seasonal design floods and operational water levels are critical for high-efficient water resource utilization. In this study, statistical and rational analyses methods were applied to divide the flood season based on seasonal rainfall patterns. The Mann–Kendall test and Theil–Sen analysis were used to detect trend changes in the observed flow series. Both stationary and nonstationary flood frequency analysis methods were conducted to estimate seasonal design floods. The Three Gorges Reservoir (TGR) in the Yangtze River, China, was selected as the case study. Results show that the TGR flood season could be divided into four periods: the reservoir drawdown period (1 May–20 June), the Meiyu flood period (21 June–31 July), the transition period (1 August–10 September), and the Autumn Rain refill period (11 September–31 October). Trend analyses indicate that the flow series at the TGR dam site exhibited a decreasing trend in recent decades. Upstream reservoir regulation has significantly reduced inflow discharges of TGR, and the nonstationary seasonal 1000-year design floods in the transition period are decreased by about 20%, and the flood control water level could rise from 145 m to 157 m, which can generate 2.288 billion kW h more hydropower (16.57% increase) while maintaining unchanged flood prevention standards. This study provides valuable insights into the TGR operational water level in the flood season and highlights the necessity of considering the regulation impact of upstream reservoirs for design floods and reservoir operational water levels. Full article

Reservoir sedimentation, a global challenge causing an annual loss of 0.8–1% of reservoir storage capacity, disrupts fluvial sediment continuity and impacts ecology and societal needs. This study focuses on the Pulangi IV reservoir in the Philippines, a shallow and wide reservoir facing significant sedimentation issues. The research aims to investigate drawdown flushing and dredging of a flushing channel for future sustainable drawdown sluicing. A test flushing event was conducted and monitoring data, including discharge, suspended sediment concentration, bathymetry, and grain size distribution, were collected. Laboratory analyses, such as critical shear stress tests, were performed for model calibration. A 3D reservoir model and a 1D sediment transport model were applied incorporating cohesive sediment behavior. Scenarios were simulated to assess drawdown flushing, dredging and downstream impacts. Results highlight the importance of drawdown level, with cohesive sediment properties playing a critical role. Sedimentation downstream of the dam, resulting from dumped or flushed sediments, was low. However, downstream ecological and morphodynamic monitoring was found to be essential for all modeled strategies. This study demonstrates potential for establishing a flushing channel enabling future sustainable drawdown sluicing during floods by conducting repeated drawdown flushing in combination with dredging in the upper reservoir. Full article

Metro foundation pits are important components of urban infrastructure projects. Dewatering and excavation are essential stages in foundation pit construction; however, this process can significantly induce groundwater drawdown, as well as diaphragm wall deformation and surface settlement. Based on a subway station foundation pit project, in this study, we employ three-dimensional numerical software to simulate the process of dewatering and excavation. A refined model is used to investigate groundwater seepage, the deformation of the retaining structure, and surface settlement under spatial effects. The finite element model accounts for stratified excavation and applied prestress conditions for the support system within the foundation pit. Its accuracy is validated through a comparison and analysis with measured data from the actual foundation pit. The results indicate that foundation pit excavation leads to significant groundwater drawdown around the pit and the formation of a characteristic “funnel-shaped” drawdown curve. Moreover, extending the diaphragm wall length contributes to maintaining a higher external groundwater level surrounding the foundation pit. The horizontal displacement of the diaphragm wall increases progressively during dewatering and excavation, and the bending moment of the diaphragm wall exhibits a trend consistent with its horizontal displacement. Surface settlement decreases as the length of the diaphragm wall increases. Full article

Reliable estimates of aquifer properties are needed for groundwater resources management and for engineering applications. Pumping tests conducted in fractured aquifers using an open borehole may not produce a proper characterization of the aquifer properties leading to the failure of engineering solutions. In this work, we apply a radial flow model to reproduce the time drawdown curves recorded at an observation borehole instrumented with multi-level piezometers drilled in the Permo-Triassic sandstone, which is a complex fractured hydraulic unit. The radial flow model and the optimization code PEST are used to estimate the aquifer hydraulic parameter values. The model is then used to investigate the implications of replacing the multi-level piezometers with an open borehole. The results show that the open borehole does not only significantly alter the groundwater head and flow patterns around the borehole, but the analysis of the time drawdown curve obtained would produce estimates of aquifer properties that bear no relationship with the actual hydraulic properties of the aquifer. For engineering control studies, the pumping test must be carefully designed to account for the presence of fractures, and these must be represented in the analysis tools to ensure the correct characterization of the hydraulic system. Full article

Soil organic carbon (SOC) mineralization is the conversion of SOC to inorganic forms of carbon (C) by microbial decomposition and conversion. It plays an important role in global C cycling. Currently, most of the studies investigating the effects of different tree species on SOC mineralization focus on forest ecosystems, and few have focused on reservoir water-level drawdown zones. In this study, we used an indoor incubation method to investigate SOC mineralization in the plantation soils of Glyptostrobus pensilis, Taxodium Zhongshanshan, Taxodium distichum and CK (unplanted plantation) in the reservoir water-level drawdown zones. We aimed to explore the effects of different tree species on the process of SOC mineralization in the reservoir water-level drawdown zones by considering both the biological and chemical processes of the soil. The results showed that the rates of SOC mineralization in the G. pensilis and T. Zhongshanshan plantations were 47% and 37%, respectively, higher than those in CK (p < 0.05), whereas the rate of SOC mineralization in T. distichum soils did not differ from that in CK. The structural equation model’s results showed microbial biomass carbon (MBC) is a key driver of SOC mineralization, while SOC and dissolved organic carbon (DOC) concentrations are also important factors that affect SOC mineralization and follow MBC. Compared to soil biochemical properties, the bacterial community composition has relatively little effect on SOC mineralization. Planted forests can, to a degree, change the biochemical properties of the soil in the reservoir water-level drawdown zones, effectively improving soil pH, and significantly increasing the amount of potential soil C mineralization, the content of SOC and the diversity of the soil bacteria (p < 0.05). Full article

With the continuous improvement of reservoir projects and the advancement of digital twin basin initiatives in China, rapidly and accurately generating long-term practical reservoir operation schedules has become a key priority for stakeholders. This study proposes a Theory-guided Long Short-Term Memory (TgLSTM) model to extract optimal reservoir operation rules accurately and reliably. Concretely, TgLSTM integrates data-fitting accuracy with the physical constraints of an operation, e.g., water level constraints and minimal discharge constraints, to address the low credibility often observed in conventional LSTM networks. Using the Three Gorges Reservoir during the dry season as a case study, a multi-year hydrological series optimized by particle swarm optimization (PSO) was used to train the TgLSTM network and derive optimized operation rules. Results show that TgLSTM efficiently generates operation schemes close to the theoretical optimum, achieving power generations of 4.27 × 10¹⁰ kW·h and 4.19 × 10¹⁰ kW·h in two test years, with deviations of only 4.20% and 2.33%, respectively. Compared to traditional LSTM models, TgLSTM is more reliable as it captures key operational characteristics such as terminal water levels and water level fluctuations, maintaining an average ten-day drawdown depth below 1.5 m—significantly lower than the 7 m fluctuations observed with conventional LSTM. Furthermore, comparative analyses against SwR, BP–ANN, and SVM confirm that TgLSTM offers a moderate performance in absolute metrics but is the best to simulate the constrained reservoir operation. Full article

In water-rich strata, a traditional vertical barrier exhibits certain limitations when applied to deep foundation pit construction under complex geological conditions, such as it is difficult to completely cut off deep and thick aquifer, which may pose potential risks during pit dewatering. To address the above challenge, this study introduced a mixed barrier system in which the horizontal barrier (HB) was set at the bottom of the foundation pit and was combined with the enclosure wall to collectively retard groundwater seepage into the pit. Based on an actual project in Tianjin, this study established HB models with varying numbers of its layers using ABAQUS 6.14 software. It systematically investigated the effect of HB on groundwater drawdown, ground surface settlement, and enclosure deflection during foundation pit dewatering. The research shows that HB can significantly reduce the magnitude of external water level drawdown by altering groundwater seepage paths while effectively controlling soil settlement. Furthermore, it exhibits favorable overall restraining effects on wall deformation. Varying the number of horizontal barrier layers (L) exhibits an insignificant effect on water-blocking and subsidence-control performance. However, the constraint effect on the enclosure shows a correlation with L. Full article

In recent years, the altered hydrological regimes and frequent extreme hydrological events in its watershed have significantly affected the stability and biodiversity of the dish-shaped lakes (DSLs) ecosystem in Poyang Lake. This study uses long-term water level records from the Xingzi hydrological station, multi-source remote sensing imagery, and field surveys to assess how altered hydrological regimes and frequent extreme hydrological events influence the coupled hydro-ecological evolution of DSLs under different gate-controlled conditions. The results reveal the following: (1) After 2003, average monthly water levels declined by 0.84 m, shifting prolonged inundation depths from the 10.0 to 14.0 m range into the 5.5 to 9.5 m range. Extreme hydrological events disrupted the hydrological regimes, triggering a clear “collapse–recovery” succession in submerged plants and major shifts in shoal wetland vegetation. (2) Gate-controlled DSLs (GC DSLs) mitigated many of these impacts by reducing the autumnal drawdown in the water area change rate to 0.324 km²/d, curbing the upward expansion of emergent and hygrophytic vegetation during high-water-level years, and stabilizing habitats during low-water-level years, although different management strategies and substrate characteristics may still lead to divergent habitat trajectories. (3) The habitat heterogeneity exhibited by the DSLs’ vegetation communities along the elevation gradient had differential effects on migratory birds, and GC DSLs can offer migratory birds relatively stable resting habitats and food resources during extreme hydrological events. The study recommends that DSL management should adopt a hierarchical dynamic regulation strategy to balance natural hydrological fluctuations with human interventions, thereby strengthening the resilience of DSL wetland habitats to extreme hydrological events. Full article

A more precise definition of groundwater dynamics is an urgent issue for developing reliable plans to assist in the sustainable management of these resources. The combination of remote sensing input data with groundwater flow models emerges as a tool capable of representing these dynamics and simulating important conditions for developing adequate groundwater exploitation plans. These technologies allow for a more detailed and accurate analysis of the interactions between the factors influencing aquifers’ behavior, such as climate variability and anthropogenic pressure on water resources. Thus, the present study aims to develop a numerical model of groundwater flow in the aquifers of the Recife Metropolitan Region, state of Pernambuco, in Brazil, to evaluate the dynamics of these waters and the piezometric level drawdowns between 2004 and 2023. The FREEWAT platform, which applies the MODFLOW-2005 code, was used to simulate the study area. The results showed the entry of seawater into some formations and drawdowns that reached more than 100 m at some points, indicating the urgent need for management strategies to mitigate salinization and preserve the quality of the region’s groundwater resources. Full article

Groundwater resources are becoming increasingly scarce, especially in arid regions of western Kazakhstan. By 2070, the domestic and drinking water demands will increase from 640 to 901 thousand m³/day. This deficiency may be overcome by utilizing the Zhem Artesian Basin’s Cretaceous Albian–Cenomanian aquifer complex. The hydrodynamic interactions between the 123 known aquifer segments and recharge zones of these promising, exploitable, high-quality groundwater sources are unclear. While MODFLOW is a nominal platform for groundwater flow assessment, which is usually used for the simulation of simple hydrological scenarios, in this study, integrating several different scales and functional modules over a GIS platform enabled delineation and the forecast of this multi-layer aquifer complex. The MODFLOW simulation assessed exploitable groundwater resources and reservoir interactions, enabling the establishment of a simultaneous operation to the Zhem aquifer complex and its neighboring reservoirs. The model suggests that the total exploitable groundwater resources may grow to 629.4 thousand m³/day during the next 50 years. The simultaneous drawdown model suggests a water level decrease of up to 80 m at the end of this period, which will cause a river flow reduction of approximately 6% of the average long-term river flow. Thus, the assessed exploitable groundwater resources will cover more than 70% of the future regional water demand. The mathematical model developed may be used for monitoring and forecasting groundwater head and water balance changes and may be applied to attain a more detailed groundwater resource transfer scheme with economic criteria. Full article

The fluctuation of reservoir water levels is a critical factor influencing the evolution of reservoir landslide–anti-slide pile systems. To investigate the reinforcement mechanism of anti-slide piles in reservoir landslides under the effect of reservoir water level fluctuations, this study employs numerical simulation methods to establish a three-dimensional slope model, simulating the drawdown process of the reservoir water level from 175 m to 145 m. The displacement and strain fields of the reservoir landslide during the water level drawdown are analyzed. Furthermore, the strain characteristics of the anti-slide pile-reinforced reservoir landslide under stress–seepage coupling are studied, and the prevention effectiveness of the landslide–anti-slide pile interaction system is explored. The results indicate that the drawdown of the reservoir water level can lead to the gradual expansion of the strain and displacement zones in the landslide, as well as a reduction in the safety factor. Under the effect of anti-slide piles, the maximum deformation of the reservoir landslide is significantly reduced. The optimal reinforcement effect is achieved when the anti-slide piles are arranged in the middle of the reservoir landslide, with a pile spacing of four times the pile diameter and an embedded depth reaching the critical depth. The findings of this study can provide a scientific basis for analyzing the instability mechanisms and mitigation of reservoir landslides. Full article

Groundwater levels were monitored before, during and after groundwater pumping to understand the impacts of groundwater extraction on Abercorn Spring, a recharge spring in the Great Artesian Basin (GAB) in southeast Queensland, Australia. We measured the wetted area of the spring during this time to understand if changes in hydrology affected the water available for vegetation communities. Sustained groundwater extraction >20 km upgradient of the spring resulted in (1) rapid drawdown of the source aquifer, causing a reduction in aquifer pressure; (2) a small decline (0.35 m) in water level at the spring; and (3) a significant change (p = 0.0001) in wetted area in winter. Recovery of water levels and wetted area of the mound spring took over three years after pumping ceased. Our study demonstrated that significant changes to the wetted area occurred with only a minimal drawdown at the springs. Abercorn Springs have a natural low variability in water level (<0.2 m), implying a stable and predictable biological community. This natural range is less than half the water level change that is currently considered for impact assessment in artesian springs in the Queensland section of the GAB, highlighting the need to incorporate updated information to inform future management of both recharge and discharge springs. In the case of Abercorn Springs, long-term monitoring and research have led to refinement of license conditions for groundwater extraction, thereby mitigating further impacts to the springs and demonstrating adaptive management. Full article

Large reservoir-induced landslides pose a persistent threat to the safety of the Three Gorges Project and the Yangtze River shipping channel. A comprehensive multi-field monitoring system has been established to observe potential landslide areas within the Three Gorges Reservoir Area. The tasks of effectively utilizing these extensive datasets and exploring the underlying correlation among various monitoring objects have become critical for understanding landslide movement patterns, assessing stability, and informing disaster prevention measures. This study focuses on the No. 1 riverside sliding mass of the Huangtupo landslide, a representative large-scale landslide in the Three Gorges Area. We specifically analyze the deformation characteristics at multiple monitoring points on the landslide surface and within underground tunnels. The analysis reveals a progressive increase in deformation rates from the rear to the front and from west to east. Representative monitoring points were selected from the front, middle, and rear sections of the landslide, along with four hydrological factors, including two reservoir water factors and two rainfall factors. These datasets were classified using the K-means clustering algorithm, while the FP-Growth algorithm was employed to uncover correlations between landslide deformation and hydrological factors. The results indicate significant spatial variability in the impacts of reservoir water levels and rainfall on the sliding mass. Specifically, reservoir water levels influence the overall deformation of the landslide, with medium-to-low water levels (146.32 to 163.23 m) or drawdowns (−18.70 to −2.16 m/month) accelerating deformation, whereas high water levels (165.37 to 175.10 m) or rising water levels (4.45 to 17.33 m/month) tend to mitigate it. In contrast, rainfall has minimal effects on the front of the landslide but significantly impacts the middle and rear areas. Given that landslide deformation is primarily driven by periodic fluctuations in reservoir water levels at the front, the movement pattern of the landslide is identified as retrogressive. The association rules derived from this study were validated using field monitoring data, demonstrating that the data mining method, in contrast to traditional statistical methods, enables the faster and more intuitive identification of reservoir-induced landslide deformation patterns and underlying mechanisms within extensive datasets. Full article