- Research Article
- 10.1016/j.watres.2026.125558
Making waves: Rethinking machine learning in wastewater effluent quality prediction through the overlooked roles of autocorrelation and baseline models.
- May 01, 2026
- Water research
- Yijie Wang + 4 more +4
Publications from 2021 to 2026
Showing 10 of 635 papers
Making waves: Rethinking machine learning in wastewater effluent quality prediction through the overlooked roles of autocorrelation and baseline models.
Comparative and integrated OMICS characterization of Italian Hericium erinaceus strains: Implications for food composition
The transition toward plant-forward diets is driving interest in sustainable, nutrient-rich food sources, with edible mushrooms emerging as promising alternatives. Hericium erinaceus mycelium, traditionally valued in Chinese medicine, is gaining recognition for its potential in food applications. This study evaluates the proteomic and metabolomic profiles of four strains of H. erinaceus mycelia collected in Italy, namely He1, He2, He4, and He5. MS proteomics and NMR metabolomics allowed the identification of 2180 proteins and 31 metabolites, revealing a rich nutritional composition. Functional analyses highlighted key pathways involved in protein synthesis and energy metabolism. Notably, strain-specific differences in amino acids, organic acids, and sugars profiles suggest opportunities for targeted strain selection to optimize nutritional profiles. These findings support the development of H. erinaceus mycelium as a sustainable, functional ingredient for novel food products, aligning with consumer demand for healthier and more environmentally conscious dietary options. • Proteomics and NMR metabolomics were applied to study Hericium erinaceus mycelium • We identified 2180 proteins and 31 metabolites across four different Italian strains • Strain-specific profiles and differences suggest targeted nutritional applications • Findings support use of H. erinaceus mycelium as a novel food ingredient
Read moreFixed-bed biofilm reactor for single-stage bioconversion of organic waste to medium-chain carboxylic acids.
A high-resolution perspective on climate drivers of lake stratification and phototrophic community dynamics in Late Glacial Central Europe
Abstract. Predicting the trajectory of aquatic deoxygenation under global warming requires a mechanistic understanding of lacustrine responses to rapid climate shifts. We investigated how climate-driven changes in catchment vegetation and local iron-rich lithology regulated lake stratification and ecosystem resilience in the maar lake Holzmaar (Central Europe). We focused on the Late Glacial, specifically on transitions during Dansgaard-Oeschger Event 1 (DOE-1; ca. 14,690–11,700 cal yr BP), a period of rapid natural warming and cooling that serves as an analogue for future high amplitude climate variation and for modern Arctic lakes undergoing rapid climate-driven transitions. Combining non-destructive hyperspectral imaging (HSI) of sedimentary pigments with high-resolution XRF geochemistry, we resolved parts of the ecosystem trajectory during DOE-1. Ecological succession progress from a pioneer community of cyanobacteria to a stable anoxic late-successional community characterized by planktonic diatom Stephanodiscus minutulus and anoxygenic purple sulphur bacteria (PSB) in the photic zone. While regional warming (mean summer temperature increased ~2.8 °C) provided the physical potential for lake stratification, our data suggest that intense anoxia was primarily triggered by the expansion of Betula in the watershed. This afforestation stabilized the water column through wind shielding. The termination of the anoxic phase coincided with the onset of the Younger Dryas cooling and increased aridity, which effectively destabilized the existing stratification. While the shift from Betula to Pinus forest may have caused a change in the terrestrial-aquatic linkage, the primary driver of the transition was the physical forcing (lake mixing) of the climatic shift (cooling). Geochemically, the lake exhibited remarkable resilience. Unlike carbonate-dominated systems prone to internal phosphorus loading, Holzmaar efficiently sequesters nutrients via a dual mechanism of reactive iron binding (authigenic vivianite) and stable mineral burial. The phosphorous trap prevents nutrient release by permanently sequestering P in the sediment, allowing rapid ecosystem recovery without delay once the specific climate and vegetation drivers shift. Our findings demonstrate that in volcanic maar lakes, catchment vegetation characteristics and local lithology can modulate, and even override, the direct effects of climate warming on aquatic anoxia.
Read moreAssessing Erosion Mitigation Effectiveness of Nature-Based Solutions Using InVEST® SDR Modeling: Application to the Carapelle Basin
Mediterranean agricultural landscapes face significant challenges from soil degradation and erosion processes that compromise both productive capacity and downstream water resources, creating an urgent need for implementing sustainable conservation strategies through Nature-Based Solutions (NBSs). This research employed the InVEST Sediment Delivery Ratio (SDR) modeling framework to examine erosional dynamics and quantify the potential benefits of various NBS interventions within the 506 km² Carapelle catchment. Model calibration and validation procedures utilized empirical sediment yield observations from the 2007-2008 monitoring period, achieving optimal parameter adjustment with only 4.3% variance from field measurements. A 20-year measured weather data were used to run the InVEST SDR model. The investigation examined four distinct NBS implementation strategies: contour-based cultivation techniques (CF), conservation tillage practices (NT), vegetative cover establishment (CCs), and integrated management approaches (Comb). Annual soil displacement rates under baseline conditions ranged between 2.43 and 3.88 t ha⁻¹ yr⁻¹ across the study years, with corresponding downstream sediment delivery of 0.86-1.30 t ha⁻¹ yr⁻¹. Conservation tillage emerged as the most effective single intervention, achieving an average 72.2% reduction in sediment transport. The integrated strategy combining conservation tillage with cover crop establishment delivered optimal results, yielding 75.9% and 70.5% reductions in sediment export and soil displacement, respectively. Geospatial evaluation demonstrated that forested and shrubland areas exhibited the highest natural retention capacity, while cultivated landscapes presented the greatest opportunities for NBSs deployment. The findings confirm that NBSs substantially improve sediment retention ecosystem services within Mediterranean agricultural watersheds. The InVEST SDR modeling approach demonstrates robust capabilities for catchment-scale erosion assessment. These outcomes offer practical insights for developing evidence-based land stewardship policies and conservation strategies in erosion-vulnerable Mediterranean regions.
Read moreFunctional diversity mapping for bumblebees to predict regional pollination service potential
To assess the extent and coverage of ecosystem services, we need to know their geographic distributions. These are typically estimated using environmental data such as land cover. But as biodiversity is closely linked with ecosystem functions, variables describing different dimensions of biodiversity should provide essential information to predict and map services produced by these functions. One persistent problem is that most of the organisms primarily responsible for services are invertebrates that suffer from data shortfalls. For example, we still lack estimates of ranges and aggregate biodiversity patterns for most described insect species. But perhaps more importantly, we are missing information on insect functional diversity, or the variation in species’ functional traits, which is most closely linked with the provision of services. The proliferation of open biodiversity data and development of robust modeling methods now make it possible to predict and map species’ distributions even for those with limited data, but the mapping of insect biodiversity patterns is just getting underway. Here, we mapped taxonomic and functional diversity for bumblebees at the country scale for Japan to assess pollination service potential. Specifically, we used citizen science data on bumblebee occurrences and species distribution models to predict bumblebee richness patterns for Japan, then used multivariate trait data to model the functional hypervolumes of bumblebee communities, and finally mapped these values to compare them with richness. We also ran extinction simulations to map functional resilience, or the ability of communities to retain functional diversity after the loss of resident species. Results showed congruences between taxonomic and functional diversity in the central Japanese Alps, also the most resilient region, but mismatch for Hokkaido in Japan’s far north with high richness yet low functional diversity. Such maps can show how dimensions of insect biodiversity differ over space and can lead to different conservation actions depending on whether species or functions are prioritized. An important next step is to integrate these results with ecosystem service models that link functional diversity estimates with actual service provision.
Read moreInelastic-stress-induced threshold instability and mobility degradation in AlN/GaN MIS-HEMTs on Si
This work investigates the impact of stress-induced effects on the stability of AlN/GaN MIS-HEMTs on Si employing PECVD-SiN as a passivation layer. DC and temperature-dependent C–V/I–V measurements together with technology computer-aided design simulations are used to correlate stress state, defect behavior, and channel transport. As-deposited PECVD-SiN introduces strong inelastic stress that compresses the 2DEG near the GaN surface and drives an anomalous reverse shift of the threshold voltage of ∼−0.3 V between 300 and 475 K, with room-temperature mobility limited to ∼700 cm2 V−1 s−1 by remote scattering. Arrhenius analysis of the drain current yields an activation energy Ea = 0.059 eV, attributed to thermally activated charging and restructuring of bulk and interface defects coupled to the stress field. Post-passivation annealing at 450 °C largely relaxes the inelastic component, broadens the potential well, and reduces Ea to 0.019 eV. The mobility increases to ∼1260 cm2 V−1 s−1 and exhibits a polar optical phonon-like temperature dependence, while the VTH–T behavior changes to a weak positive coefficient. These results offer insights on the temperature stability of stress-induced GaN MIS-HEMTs from a material perspective.
Read moreEngineering Electro-microbial Routes for Carbon-Neutral Biomanufacturing
Climate-eutrophication-anoxia interactions in Late Glacial Soppensee, Switzerland: Forcings, non-linear responses and recovery
Combined effects of climate warming and anthropogenic nutrient loadings lead to lake eutrophication and anoxia globally. Because of chemical feedbacks, lakes under multiple stressors often respond in non-linear ways. However, it remains unclear whether climate change alone can lead to non-linear lake responses in the absence of anthropogenic nutrient disturbances. Here, we investigate the interactions between climate variability, nutrient cycling and trophic state changes, mixing regimes, anoxia and related chemical feedback in a small kettle-hole lake in Switzerland during Late Glacial times (15.2–12.6 cal ka BP), a period known for high-amplitude climate change in pre-anthropogenic times. After its formation during Heinrich Stadial 1 (>15 cal ka BP), Soppensee was oligotrophic and well-mixed. Soppensee became eutrophic and developed anoxia at 14.25 cal ka BP. Phosphorus (P) was released from sediments through the reductive dissolution of Fe-oxyhydroxides, fuelling eutrophication. Eutrophication lagged the Bølling warming (14.65 cal ka BP) by 400 years, suggesting that rising temperatures were not the trigger for eutrophication. Instead, eutrophication responded non-linearly to forest closure (threshold at 76 % arboreal pollen AP), which shielded Soppensee from wind mixing, enhancing lake stratification, anoxia and P release, intensifying eutrophication. These conditions ended during the 200-years cold period of the Aegelsee Oscillation (GI-1d, ca. 14.0 cal ka BP) when the landscape regionally opened (AP<76 %); the lake became well-mixed, oxygenated and P was efficiently sequestered. Throughout the Allerød (13.9–12.8 cal ka BP), enhanced Fe input prompted diagenetic vivianite formation, sequestering P in sediments, naturally remediating lake eutrophication despite closed forests, warm temperatures, lake stratification and anoxia. • Soppensee experienced natural eutrophication and anoxia during the Bølling. • Eutrophication lagged warming by 400 years and showed a hysteresis response to closed forest cover. • Closed forests reduced wind mixing, enhanced thermal stratification, hypolimnetic anoxia and internal P release. • Fe-P (vivianite) sequestration during the Allerød caused natural oligotrophication despite warm climate and closed forests.
Read moreEnhancing water quality assessment in Skikda, Algeria using the PCA-based weighted index (WQI_P) and its predictive performance: a comparison with traditional WA_WQI approaches.
Ensuring reliable river-water quality assessment is increasingly important in North Africa, where pollution pressures and data limitations complicate monitoring. Therefore, the research developed a principal-component-analysis-based water quality index (WQI_P) that is designed to address eclipsing, multicollinearity, and subjectively assigned weights that affect traditional indices such as the weighted-arithmetic WQI (WA_WQI). The objective of the research is to evaluate whether PCA-derived weights and objective parameter selection improve reliability, uncertainty, and classification stability. A dataset of 159 river-water samples from the Skikda region (Algeria) was analyzed. After screening correlated variables and extracting PCA contributions, WQI_P was constructed from the retained components. Eight machine-learning algorithms and a stacked ensemble were used under 10-fold cross-validation to compare the prediction performance and uncertainty of WQI_P and WA_WQI. Agreement metrics, PREI scores, confidence intervals, and class-transition analysis were used to assess the differences between the two indices, Predictive uncertainty was quantified using a Gaussian Monte Carlo simulation, which propagates variability by repeatedly perturbing model residuals to generate distributions of index predictions. The WQI_P consistently produced lower prediction errors (stacked RMSE=2.74; MAE=1.75) than the WA_WQI (RMSE=3.16; MAE=2.21), together with narrower 95% confidence intervals and reduced predictive uncertainty. The classification outcomes shifted toward a stricter and more balanced assessment: the proportion of samples classified as "Excellent" decreased (30 to 7), "Good" increased (55 to 88), and "Unsuitable" declined (40 to 12). These results indicated that grounding weights in the multivariate structure enhances stability and reduces dependence on a small set of dominant parameters. The findings demonstrated that the WQI_P can improve transparency, objectivity, and monitoring efficiency by focusing on the most informative variables. The index is applicable to data-scarce regions where objective weighting and uncertainty control are essential. Future work should test WQI_P across larger and more heterogeneous basins, extend validation using spatial-temporal blocking, and explore its integration into operational monitoring frameworks.
Read more