Reassessing the relationship between landscape alteration and aquatic ecosystem degradation from a hydrologically sensitive area perspective

Prioritizing ecological restoration in hydrologically sensitive areas to improve groundwater quality

Greening is the optimal way to mitigate climate change and water quality degradation caused by agricultural expansion and rapid urbanization. However, the ideal sites to plant trees or grass to achieve a win-win solution between the environment and the economy remain unknown. Here, we performed a nationwide survey on groundwater nutrients (nitrate nitrogen, ammonia nitrogen, dissolved reactive phosphorus) and heavy metals (vanadium, chromium, manganese, iron, cobalt, nickel, copper, arsenic, strontium, molybdenum, cadmium, and lead) in China, and combined it with the global/national soil property database and machine learning (random forest) methods to explore the linkages between land use within hydrologically sensitive areas (HSAs) and groundwater quality from the perspective of hydrological connectivity. We found that HSAs occupy approximately 20 % of the total land area and are hotspots for transferring nutrients and heavy metals from the land surface to the saturated zone. In particular, the proportion of natural lands within HSAs significantly contributes 8.0 % of the variability in groundwater nutrients and heavy metals in China (p < 0.01), which is equivalent to their contribution (8.8 %) at the regional scale (radius = 4 km, area = 50 km2). Increasing the proportion of natural lands within HSAs improves groundwater quality, as indicated by the significant reduction in the concentrations of nitrate nitrogen, manganese, arsenic, strontium, and molybdenum (p < 0.05). These new findings suggest that prioritizing ecological restoration in HSAs is conducive to achieving the harmony between the environment (improving groundwater quality) and economy (reducing investment in area management).

Monthly runoff prediction by a multivariate hybrid model based on decomposition-normality and Lasso regression

The intensified non-stationary, skewness, non-linear nature of runoff series due to the comprehensive influences of meteorological events and human activities has brought new challenges to accurate runoff prediction. To solve the issues, a multivariate hybrid model introducing decomposition-normality mode into SVR was proposed. The normal transformation techniques, Box-Cox transformation, and W-H inverse transformation were employed to transform the input variables of the model into normal distribution to overcome the error caused by skewness of the runoff data. The results show that decomposition-normality mode can improve the performance of the models. In particular, WT-BC-LSVR accurately predicted peak flow and low flow during the testing, and the mean relative errors are less than 16%, Rs and Nash-Sutcliffe efficiencies are greater than 0.97 and 0.94, respectively. The study demonstrates that the proposed multivariate hybrid model based on the decomposition-normality mode is a novel promising prediction model with satisfactory performance that can accurately predict complex monthly runoff.

Hydrological connectivity affects nitrogen migration and retention in the land‒river continuum

Land use change and excessive nitrogen (N) loading threaten the health of receiving water bodies worldwide. However, the role of hydrological connectivity in linking watershed land use, N biogeochemistry and river water quality remain unclear. In this study, we investigated 15 subwatersheds in the Jiulong River watershed (southeastern China) during a dry baseflow period in 2018, combined with 3‒year (2017-2019) nutrient monitoring in 5 subwatersheds to explore river N dynamics (dissolved nutrients, dissolved gases and functional genes) and their controlling factors at three hydrological connectivity scales, i.e., watershed, hydrologically sensitive areas (HSAs) and riparian zone. The results show that land use at HSAs (less than 20% of watershed area) and watershed scales contributed similarly to river N variation, indicating that HSAs are hotspots for transporting land N into river channels. In particular, the agricultural land was the main factor affecting river nitrate and nitrous oxide (N₂O) concentrations, while the built-up land significantly affected river ammonium and nitrite. At the riparian zone scale, soils and sediments substantially influenced river N retention processes (i.e., nitrification and denitrification). Management and protection measures targeting HSAs and riparian zones are expected to efficiently reduce river N loading and improve water quality.

Water quality analysis based on phytoplankton and metal indices: a case study in the Sauce Grande River Basin (Argentina)

The increasing landscape alterations due to anthropogenic activities is of global concern since it affects aquatic ecosystems, often resulting in compromise of the ecological integrity and the water quality. In this sense, the evaluation, monitoring, and prediction of the aquatic ecosystem quality becomes an important research subject. This study presents the first integrated water quality assessment of the Sauce Grande River Basin, in Argentina, based on the spatial distribution of the phytoplankton community, the physicochemical parameters, and the metal concentrations (Cd, Cu, Cr, Fe, Mn, Ni, Pb, and Zn) found in the particulate fraction. According to the trophic indices and the phytoplankton abundance, composition, and diversity, the water quality showed significant deterioration in the lower basin after the Sauce Grande lake. The trophic state index indicated that water was oligotrophic in over 75% of the sampling sites, increasing downstream, where two sites were characterized as mesotrophic, and one described as hypertrophic. The phytoplankton community was dominated by diatoms in zones with low anthropogenic impact and conductivity, whereas high densities of Euglenophyta, Chlorophyta, and Cyanobacteria were found in the middle-lower basin, associated with higher organic matter and eutrophication. The conductivity, turbidity, and most metal concentrations also increased towards the downstream area, even exceeding recommended levels for the metals Cu, Cr, Mn, and Pb in the middle and lower reaches of the basin (Cu: 3.5 µg L^-1; Cr: 2.4 µg L^-1; Pb: 1.2 µg L^-1; Mn 170 µg L^-1). This study generates a database for the water quality of the Sauce Grande River Basin and sets an example of how the water quality varies along a basin that crosses different topographic environments, land covers, and anthropogenic influences.

Anthropogenic factors associated with contaminants of emerging concern detected in inland Minnesota lakes (Phase II)

Contaminants of emerging concern (CECs) include a variety of pharmaceuticals, personal care products, and hormones commonly detected in surface waters. Human activities, such as wastewater treatment and discharge, contribute to the distribution of CECs in water, but other sources and pathways are less frequently examined. This study aimed to identify anthropogenic activities and environmental characteristics associated with the presence of CECs, previously determined to be of high priority for further research and mitigation, in rural inland lakes in northeastern Minnesota, United States. The setting for this study consisted of 21 lakes located within both the Grand Portage Indian Reservation and the 1854 Ceded Territory, where subsistence hunting and fishing are important to the cultural heritage of the indigenous community. We used data pertaining to numbers of buildings, healthcare facilities, wastewater treatment plants, impervious surfaces, and wetlands within defined areas surrounding the lakes as potential predictors of the detection of high priority CECs in water, sediment, and fish. Separate models were run for each contaminant detected in each sample media. We used least absolute shrinkage and selection operator (LASSO) models to account for both predictor selection and parameter estimation for CEC detection. Across contaminants and sample media, the percentage of impervious surface was consistently positively associated with CEC detection. Number of buildings in the surrounding area was often negatively associated with CEC detection, though nonsignificant. Surrounding population, presence of wastewater treatment facilities, and percentage of wetlands in surrounding areas were positively, but inconsistently, associated with CECs, while catchment area and healthcare centers were generally not associated. The results of this study highlight human activities and environmental characteristics associated with CEC presence in a rural area, informing future work regarding specific sources and transport pathways. We also demonstrate the utility of LASSO modeling in the identification of these important relationships.

Water quality prospective in Twenty First Century: Status of water quality in major river basins, contemporary strategies and impediments: A review

Water quality improvement is one of the top priorities in the global agenda endorsed by United Nation. In this review manuscript, a holistic view of water quality degradation such as concerned pollutants, source of pollution, and its consequences in major river basins around the globe (at least 1 from each continent and a total of 16 basins) is presented. Additionally, nine contemporary techniques such as field scale evaluation, watershed scale evaluation, strategies to identify critical source areas, optimization strategies for placement of best management practices (BMPs), social component in watershed modeling, machine learning algorithms to address water quality problems in complex natural systems concomitant with spatial heterogeneity, establishing a total maximum daily loads (TMDLs), remote sensing in monitoring water quality, and developing water quality index are discussed. Next, the existing barriers to improve water quality are classified into primary and secondary impediments. A detail discussion of three primary impediments (climate change, urbanization and industrial activities, and agriculture) and ten secondary impediments (availability of water quality data, complexity of system, lack of skilled person, environmental legislation, fragmented mandate, limitation in resources, environmental awareness, resistance to change, alteration of nutrient ratio by river damming, and emerging pollutants) are illustrated. Finally, considering all the existing knowledge gaps pertaining to contemporary strategies, a future direction of water quality research is outlined to significantly improve the water quality around the globe.

Linking Coleopteran Diversity With Agricultural Management of Maize-Based Agroecosystems in Oaxaca, Mexico

Biodiversity is known to be influenced by agricultural practices in many ways. However, it is necessary to understand how this relation takes place in particular agroecosystems, sociocultural contexts and for specific biological groups, especially in highly biodiverse places. Also, in order to systematically study and track how biodiversity responds or changes with agricultural practices, it is necessary to find groups that can be used as practical indicators. We conduct a study of beetle (Coleoptera) diversity in maize-based agricultural plots with heterogeneous management practices in the Central Valleys of Oaxaca, Mexico, a region with outstanding biodiversity and a long agricultural history. We use a mixture of local knowledge and multivariate statistics to group the plots into two broad and contrasting management categories (traditional vs. industrialized). Then, we present an analysis of Coleopteran diversity for each category, showing higher levels across different diversity indexes for the traditional plots. Specifically, Coleopteran guilds associated with natural pest control and soil conservation are more common in traditional plots than in industrialized ones, while herbivorous beetles are more abundant in the second. Also, our results let us postulate the Curculionidae family as an indicator of both management type and overall Coleopteran diversity in the agricultural lands of the study site. We discuss our results in terms of the agricultural matrix quality and its role in strategies that favor the coexistence of culturally meaningful agricultural systems and local biodiversity.

Quantifying climate, streamflow, and watershed control on water quality across Southeastern US watersheds

Identification of the key variables that influence spatial variation in stream water quality is crucial for designing sustainable water management strategies. In this study, we investigated the key variables that influence the spatial variability of stream of water quality, across multiple watersheds. This study uses water quality data collected over 19 years for 59 watersheds located in the Southeast Atlantic region of the United States, which includes the states of North Carolina, South Carolina, and Georgia. A conceptual modeling framework was developed to understand the linkage between the long-term mean water quality constituents (Total nitrogen, Total phosphorus, Turbidity, and pH) and the watershed characteristics (e.g., topography, land use/cover, soil type), streamflow data, and climatic variables (precipitation and temperature). The modeling results suggest that not only anthropogenic variables influence the mean water quality constituents, but other watershed characteristics, such as soil properties, have a significant impact. The natural watershed characteristics explain most of the spatial variability in the mean Turbidity and pH values in streams. The modeling results also suggest that once land use and soil properties are considered, watershed topography has a limited role to explain the variation in the mean water quality. Overall, the developed watershed models can be used to forecast stream water-quality responses to future land use, climate, soil, and land management changes within the study area.

Bacterioplankton community responses and the potential ecological thresholds along disturbance gradients

Increasing intensity and frequency of coastal pollutions are the trajectory to be expected due to anthropogenic pressures. However, it is still unclear how and to what extent bacterioplankton communities respond to the two factors, despite the functional importance of bacterioplankton in biogeochemical cycles. In this study, significant organic pollution index (OPI) and offshore distance gradients, as respective proxies of disturbance intensity and disturbance frequency, were detected in a regional scale across the East China Sea. A multiple regression on matrices (MRM) revealed that the biogeography of bacterioplankton community depended on spatial scale, which was governed by local characters. Bacterioplankton community compositions (BCCs) were primarily governed by the conjointly direct (-0.28) and indirect (-0.48) effects of OPI, while offshore distance contributed a large indirectly effect (0.52). A SEGMENTED analysis depicted non-linear responses of BCCs to increasing disturbance intensity and disturbance frequency, as evidenced by significant tipping points. This was also true for the dominant bacterial phyla. Notably, we screened 30 OPI-discriminatory taxa that could quantitatively diagnose coastal OPI levels, with an overall 79.3% accuracy. Collectively, the buffer capacity of bacterioplankton communities to increasing disturbance intensity and disturbance frequency is limited, of which the significant tipping points afford a warning line for coastal management. In addition, coastal pollution level can be accurately diagnosed by a few OPI-discriminatory taxa.

Evaluating the impact of land uses on stream integrity using machine learning algorithms

A general pattern of declining aquatic ecological integrity with increasing urban land use has been well established for a number of watersheds worldwide. A more nuanced characterization of the influence of different urban land uses and the determination of cumulative thresholds will further inform watershed planning and management. To this end, we investigated the utility of two machine learning algorithms (Random Forests (RF) and Boosted Regression Trees (BRT)) to model stream impairment through multimetric macroinvertebrate index known as High Gradient Macroinvertebrate Index (HGMI) in an urbanizing watershed located in north-central New Jersey, United States. These machine learning algorithms were able to explain at least 50% of the variability of stream integrity based on watershed land use/land cover. While comparable in results, RF was found to be easier to train and was somewhat more robust to model overfitting compared to BRT. Our results document the influence of increasing high-medium density (> 30% Impervious Surface cover (ISC)), low density (15-30% ISC) urban and transitional/barren land had in negatively affecting stream biological integrity. The thresholds generated by partial plots suggest that the stream integrity decreased abruptly when the percentage of high-medium and low density urban, and transitional/barren land went above 10%, 8%, and 2% of the watershed, respectively. Additionally, when rural residential surpassed 30% threshold, it behaved similar to low density urban towards stream integrity. Identification of such cumulative thresholds can help watershed managers and policymakers to craft land use zoning regulations and design restoration programs that are grounded by objective scientific criteria.

The assessment of water erosion using Partial Least Squares-Path Modeling: A study in a legally protected area with environmental land use conflicts

Water erosion has historically been assessed by various methods, with the purpose to help reducing this phenomenon. However, application of models capable to handle complex relationships between large numbers of variables is still relatively scarce. The method of Partial Least Squares-Path Modeling (PLS-PM), used in this study, was able to expose complex causal paths between soil erosion and potentially related factors, namely "Surface Runoff", "Environmental Land Use Conflicts", "Soil Fertility" and "Relief Factors", within the Environmental Protection Area of Uberaba River Basin (EPA) located in Minas Gerais state, Brazil. In the context of PLS-PM, soil erosion (dependent) and the related factors (independent) are called latent variables and described by measured or estimated parameters. For example, the "Relief Factors" were described by measured drainage density and topographic slope. These were linked to the corresponding latent variables through weights and the later joined to each other through paths. During the PLS-PM runs, weights and paths were quantified and latent variables interpreted in regard to their importance for soil erosion and spatial incidence. The spatial incidence was used to prioritize areas for soil conservation. To test the model, data were obtained from soil samples (texture and fertility parameters) or digitally extracted from cartographic products (e.g., maps of soil loss, land use, brightness index, topographic slope, drainage density), at 37 sites within the EPA. The PLS-PM results revealed that 70.2% of soil erosion is predicted by the independent variables (R² = 0.702), and that "Soil Fertility" and "Environmental Land Use Conflicts" were the most influencing ones (β = -0.758 and β = 0.346, respectively). These variables can be managed by man, through implementation of effective soil conservation measures and respect for suitable land use. It is therefore urgent to act in these regard, considering the socioeconomic and environmental importance of the EPA.