Pollution loads in the middle-lower Yangtze river by coupling water quality models with machine learning.

Pollution control and environmental protection of the Yangtze River have received major attention in China. However, modeling the river's pollution load remains challenging due to limited monitoring and unclear spatiotemporal distribution of pollution sources. Specifically, anthropogenic activities' contribution to the pollution have been underestimated in previous research. Here, we coupled a hydrodynamic-based water quality (HWQ) model with a machine learning (ML) model, namely attention-based Gated Recurrent Unit, to decipher the daily pollution loads (i.e., chemical oxygen demand, COD; total phosphorus, TP) and their sources in the Middle-Lower Yangtze River from 2014 to 2018. The coupled HWQ-ML model outperformed the standalone ML model with KGE values ranging 0.77-0.91 for COD and 0.47-0.64 for TP, while also reducing parameter uncertainty. When examining the relative contributions at the Middle Yangtze River Hankou cross-section, we observed that the main stream and tributaries, lateral anthropogenic discharges, and parameter uncertainty contributed 15, 66, and 19% to COD, and 58, 35, and 7% to TP, respectively. For the Lower Yangtze River Datong cross-section, the contributions were 6, 69, and 25% for COD and 41, 42, and 17% for TP. According to the attention weights of the coupled model, the primary drivers of lateral anthropogenic pollution sources, in descending order of importance, were temperature, date, and precipitation, reflecting seasonal pollution discharge, industrial effluent, and first flush effect and combined sewer overflows, respectively. This study emphasizes the synergy between physical modeling and machine learning, offering new insights into pollution load dynamics in the Yangtze River.

Impacts of urban expansion on air temperature and humidity during 2022 mega-heatwave over the Yangtze River Delta, China.

The Yangtze River Delta (YRD) experienced record-breaking heat in the summer of 2022. However, the urban heat pattern and the role of urban expansion over the last two decades in this hot summer have not been explored. Using the advanced mesoscale Weather Research and Forecasting (WRF) model, we reproduced the fine spatial features and investigated the urban heat island (UHI) and dry island (UDI) effects during the two identified heatwaves in 2022. We further replace the current (2020) land use with the historical (2001) land use in WRF to evaluate the impacts of urban expansion from 2001 to 2020 on air temperature and moisture. Our finding revealed that the conversion of land use resulted in near-surface warming and drying, with pronounced diurnal variations, especially during the July heatwave. The analysis of surface energy balance demonstrated that the substantial decrease in evapotranspiration (ET) was the primary driver of daytime warming, elevating temperatures by 7 °C (July heatwave) and 2 °C (August heatwave). This ET reduction also led to the strong daytime coupling of warming and drying effects over new urban areas. At night, the release of stored heat resulted in the temperature increase of 2 °C (1 °C) during July (August) heatwave, highlighting the nighttime as a critical period for heightened thermal risk. Additionally, urban expansion at the periphery contributed modestly to the warming of urban cores, exacerbating conditions in an already hot environment. This study enhances understanding of the impacts of urban expansion on air temperature and humidity during extreme heatwaves, thereby supporting targeted adaptation and mitigation for extreme events within large cities.

Coupled natural-social water system in a subwatershed of the Tennessee River Basin.

Regional water cycle systems are increasingly characterized by the dual effect of natural and social processes, which have profound impacts on global water security. However, accurately interpreting the changes in the coupled natural-social water system and identifying the driving factors pose significant challenges. Here, we attempted to model a coupled natural-social water system in the East Fork Poplar Creek (EFPC) watershed of the Tennessee River, United States. The study area features two social water cycle components: a local water transfer project and the Oak Ridge Wastewater Treatment Facility (ORWTF). We conducted the Soil and Water Assessment Tool (SWAT) modeling in the open-source light-weight QGIS software, with the synthesis of various climate and land use change scenarios in both historical periods (1980-2016) and future periods (2017-2050). We achieved more accurate and realistic model simulations when considering the social water cycle components, indicating that the social water cycle accounted for 13-18 % of the observed streamflow. Climate variation/change dominates natural runoff changes. Though land use and cover change (LUCC) had minimal effect on natural runoff, it had a profound impact on the process of runoff generation, i.e., surface runoff (RS) and subsurface runoff (RSS). Specifically, LUCC would be responsible for 152 % and 45 % of the changes in RS and RSS, respectively, in future periods. This study highlights the significance of artificial water discharge and withdrawal impacts on the water cycle and emphasizes the need for water resources management measures that fully consider natural-social hydrological processes.

Rainfall effect on soil respiration depends on antecedent soil moisture.

Climate change has a discernible influence on rainfall patterns, thus potentially affecting the intricate dynamics of soil respiration (Rs) and soil carbon storage. However, we still lack a profound understanding of the determinants of Rs response to rainfall events. Here, utilizing a comprehensive 10-year dataset (2004-2013), we explored the direction and magnitude of Rs response to rainfall events and the underlying determinants in a temperate forest. Based on the identified 368 rainfall events over the study period, we demonstrate that rainfall suppresses Rs when the soil moisture is optimal and moist in the growing season, whereas its effect on Rs during the non-growing season is minimal. Notably, antecedent soil moisture, rather than rainfall amount, shows a substantial impact on Rs during the growing season (coefficient of determination (R2) = 0.37 for antecedent soil moisture, and R2 < 0.01 for rainfall amount). Incorporating antecedent soil moisture significantly enhances the explanatory power (R2) from 0.09 to 0.45 regarding the relative changes in Rs following rainfall events. Our results highlight the environmental dependency of Rs response to rainfall events and suggest that incorporating the role of antecedent soil moisture could enhance predictability and reduce uncertainty in ecosystem modeling.

Water quality prediction based on sparse dataset using enhanced machine learning.

Water quality in surface bodies remains a pressing issue worldwide. While some regions have rich water quality data, less attention is given to areas that lack sufficient data. Therefore, it is crucial to explore novel ways of managing source-oriented surface water pollution in scenarios with infrequent data collection such as weekly or monthly. Here we showed sparse-dataset-based prediction of water pollution using machine learning. We investigated the efficacy of a traditional Recurrent Neural Network alongside three Long Short-Term Memory (LSTM) models, integrated with the Load Estimator (LOADEST). The research was conducted at a river-lake confluence, an area with intricate hydrological patterns. We found that the Self-Attentive LSTM (SA-LSTM) model outperformed the other three machine learning models in predicting water quality, achieving Nash-Sutcliffe Efficiency (NSE) scores of 0.71 for CODMn and 0.57 for NH3N when utilizing LOADEST-augmented water quality data (referred to as the SA-LSTM-LOADEST model). The SA-LSTM-LOADEST model improved upon the standalone SA-LSTM model by reducing the Root Mean Square Error (RMSE) by 24.6% for CODMn and 21.3% for NH3N. Furthermore, the model maintained its predictive accuracy when data collection intervals were extended from weekly to monthly. Additionally, the SA-LSTM-LOADEST model demonstrated the capability to forecast pollution loads up to ten days in advance. This study shows promise for improving water quality modeling in regions with limited monitoring capabilities.

Effect of Cinnamaldehyde on Systemic Candida albicans Infection in Mice.

OBJECTIVE: To investigate the therapeutic efficacy of cinnamaldehyde (CA) on systemic Candida albicans infection in mice and to provide supportive data for the development of novel antifungal drugs. METHODS: Ninety BALB/c mice were randomly divided into 3 groups according to a random number table: CA treatment group, fluconazole (positive control) group, and Tween saline (negative control) group, with 30 mice in each group. Initially, all groups of mice received consecutive intraperitoneal injections of cyclophosphamide at 200 mg/kg for 2 days, followed by intraperitoneal injection of 0.25 mL C. albicans fungal suspension (concentration of 1.0 × 107 CFU/mL) on the 4th day, to establish an immunosuppressed systemic Candida albicans infection animal model. Subsequently, the mice were orally administered CA, fluconazole and Tween saline, at 240, 240 mg/kg and 0.25 mL/kg respectively for 14 days. After a 48-h discontinuation of treatment, the liver, small intestine, and kidney tissues of mice were collected for fungal direct microscopic examination, culture, and histopathological examination. Additionally, renal tissues from each group of mice were collected for (1,3)- ß -D-glucan detection. The survival status of mice in all groups was monitored for 14 days of drug administration. RESULTS: The CA group exhibited a fungal clearance rate of C. albicans above 86.7% (26/30), significantly higher than the fluconazole group (60.0%, 18/30, P<0.01) and the Tween saline group (30.0%, 9/30, P<0.01). Furthermore, histopathological examination in the CA group revealed the disappearance of inflammatory cells and near-normal restoration of tissue structure. The (1,3)-ß-D-glucan detection value in the CA group (860.55 ± 126.73 pg/mL) was significantly lower than that in the fluconazole group (1985.13 ± 203.56 pg/mL, P<0.01) and the Tween saline group (5910.20 ± 320.56 pg/mL, P<0.01). The mouse survival rate reached 90.0% (27/30), higher than the fluconazole group (60.0%, 18/30) and the Tween saline group (30.0%, 9/30), with a significant difference between the two groups (both P<0.01). CONCLUSIONS: CA treatment exhibited significant therapeutic efficacy in mice with systemic C. albicans infection. Therefore, CA holds potential as a novel antifungal agent for targeted treatment of C. albicans infection.

Experimental warming accelerates positive soil priming in a temperate grassland ecosystem.

Unravelling biosphere feedback mechanisms is crucial for predicting the impacts of global warming. Soil priming, an effect of fresh plant-derived carbon (C) on native soil organic carbon (SOC) decomposition, is a key feedback mechanism that could release large amounts of soil C into the atmosphere. However, the impacts of climate warming on soil priming remain elusive. Here, we show that experimental warming accelerates soil priming by 12.7% in a temperate grassland. Warming alters bacterial communities, with 38% of unique active phylotypes detected under warming. The functional genes essential for soil C decomposition are also stimulated, which could be linked to priming effects. We incorporate lab-derived information into an ecosystem model showing that model parameter uncertainty can be reduced by 32-37%. Model simulations from 2010 to 2016 indicate an increase in soil C decomposition under warming, with a 9.1% rise in priming-induced CO2 emissions. If our findings can be generalized to other ecosystems over an extended period of time, soil priming could play an important role in terrestrial C cycle feedbacks and climate change.

Exploring the competitive dynamic enzyme allocation scheme through enzyme cost minimization.

Enzyme allocation (or synthesis) is a crucial microbial trait that mediates soil biogeochemical cycles and their responses to climate change. However, few microbial ecological models address this trait, particularly concerning multiple enzyme functional groups that regulate complex biogeochemical processes. Here, we aim to fill this gap by developing a COmpetitive Dynamic Enzyme ALlocation (CODEAL) scheme for six enzyme groups that act as indicators of inorganic nitrogen (N) transformations in the Microbial-ENzyme Decomposition (MEND) model. This allocation scheme employs time-variant allocation coefficients for each enzyme group, fostering mutual competition among the multiple groups. We show that the principle of enzyme cost minimization is achieved by using the substrate's saturation level as the factor for enzyme allocation, resulting in an enzyme-efficient pathway with minimal enzyme cost per unit metabolic flux. It suggests that the relative substrate availability affects the trade-off between enzyme production and metabolic flux. Our research has the potential to give insights into the nuanced dynamics of the N cycle and inspire the evolving landscape of enzyme-mediated biogeochemical processes in microbial ecological modeling, which is gaining increasing attention.

Ecosystem services from partially harvested riparian buffers can offset biomass production costs.

There is a broad consensus that riparian buffers provide environmental benefits and increase resilience to climate change. In this study, we examined the potential benefits of multi-zone riparian buffers with outer layers planted in perennial crops (i.e., partially harvested buffers). This was accomplished by developing a simplified regional modeling tool, BioVEST, which was applied in the Mid-Atlantic region of the USA. Our analysis revealed that a substantial portion of variable costs to produce biomass for energy can potentially be offset by values provided by ecosystem services from partially harvested riparian buffers. Ecosystem services were monetized and found to represent a substantial fraction (median = ~42%) of variable crop production cost. Simulated water-quality improvements and carbon benefits generally occurred where buffer area was available, but hotspots occurred in different watersheds, suggesting potential trade-offs in decisions about buffer locations. A portion of buffers could be eligible for ecosystem service payments under US government incentive programs. Partially harvested buffers could represent a sustainable and climate-resilient part of multi-functional agricultural landscapes, and one that could become economically viable if farmers are able to reap the value of providing ecosystem services and if logistical challenges are resolved. Our results suggest that payments for ecosystem services can close the gap between what biorefineries are willing to pay and what landowners are willing to accept to grow and harvest perennials along streams.

Global patterns and edaphic-climatic controls of soil carbon decomposition kinetics predicted from incubation experiments.

Knowledge about global patterns of the decomposition kinetics of distinct soil organic matter (SOM) pools is crucial to robust estimates of land-atmosphere carbon fluxes under climate change. However, the current Earth system models often adopt globally-consistent reference SOM decomposition rates (kref), ignoring effects from edaphic-climate heterogeneity. Here, we compile a comprehensive set of edaphic-climatic and SOM decomposition data from published incubation experiments and employ machine-learning techniques to develop models capable of predicting the expected sizes and kref of multiple SOM pools (fast, slow, and passive). We show that soil texture dominates the turnover of the fast pools, whereas pH predominantly regulates passive SOM decomposition. This suggests that pH-sensitive bacterial decomposers might have larger effects on stable SOM decomposition than previously believed. Using these predictive models, we provide a 1-km resolution global-scale dataset of the sizes and kref of these SOM pools, which may improve global biogeochemical model parameterization and predictions.