Runoff Events Research Articles

Nested hydrological modeling for flood prediction using CMIP6 inputs around Lake Tana, Ethiopia

Study regionTana basin is the origin of the Blue Nile Basin located in Ethiopia. The Lake Tana’s mean annual precipitation is approximately 1400 mm/yr and its outflow drains an area of about 15,096 km2. There is limited effort to apply nested hydrological models for flood prediction due to being poorly gauged for validation. Study focusThe objective of this study is to show how climate simulations can be used to generate reliable local predictions of flood and runoff for the Blue Nile in the Ethiopian highlands.In this study, outputs from the Coupled Model Intercomparison Project Phase 6 (CMIP6) were used to initialize a nested hydrological model to reveal long-term trends of runoff in flood-prone areas of the Lake Tana basin, Ethiopia. Available satellite and reanalysis datasets were used for selection of CMIP6 products and the five models with best validation were used to force a nested hydrological model: Known as Wflow_sbm. A model-independent multi-algorithm optimization estimation tool was implemented for calibration of Wflow with insitu observations. New hydrological insights for the regionIn terms of simulating runoff and flood events, application of Wflow_sbm to the Lake Tana basin gave promising results. This study serves as a major step towards the development and implementation of global model-driven nested hydrological assessments of flood risk in future projections for Lake Tana basin.

A novel ensemble algorithm based on hydrological event diversity for urban rainfall–runoff model calibration and validation

Hydrological models such as the stormwater management model (SWMM) are essential tools for managing water resources. Model performance typically relies on calibration to historic rainfall–runoff events. However, traditional calibration techniques suffer many issues, such as ad-hoc event selection, ignoring model parameter uncertainty, and poor validation techniques. We introduce a novel ensemble framework that relies on data-driven event selection of calibration and validation events to achieve robust model generalisation.The proposed framework is applied to SWMM models for two urbanised catchments in Southern Ontario, Canada. The framework uses unsupervised machine learning to cluster rainfall–runoff events. Events in different clusters have diverse characteristics; sampling events from each cluster ensures a diverse set of events for calibration. Validation events can be sampled in the same manner, which ensures similarity between calibration and validation data. We study the effects that the calibration event cluster has on model sensitivity and generalisation. Only minor differences in model parameter sensitivity were observed between different clusters. However, the cluster(s) used for calibration were found to be very important for model generalisation. Next, we generate ensemble predictions, which use multiple SWMM models, each calibrated to different events. Ensemble performance is compared to the single-model baseline and across increasing numbers of ensemble members. Results demonstrate improvements in generalisation and decrease in prediction variance across increasing ensemble size. Finally, we compare several different techniques for ensemble combination, which increase in complexity from simple averaging to ANN-based stacking augmented with exogenous input features. The ANN-based stacking performs the best according to two out of three evaluation criteria. Simple averaging, while inferior in performance to stacking, consistently outperforms the single model baseline. The novel framework proposed in this work provides a data-driven, adaptable alternative to conventional calibration frameworks that have been proven to improve model performance.

The response of freshwater plankton communities to temporal concurrence of agrochemical mixtures

Freshwater ecosystems regularly experience pulsed inputs of nutrients and other pollutants as a result of temporally variable applications of agrochemicals combined with runoff events. In this study, we evaluate how planktonic communities respond to repeated and pulsed insecticide disturbances and if the response depends on temporal concurrence with nutrient pulses. We conducted an experiment using mesocosms to assess the ecotoxicological effects of a commonly used insecticide (chlorpyrifos) and nutrients (nitrogen and phosphorus) on plankton communities. The mesocosms (300 L) were established outdoors for 10 weeks. The experiment consisted of 3 treatments: nutrient pulse of nitrogen and phosphorus every two weeks (N; considered as control), nutrients and insecticide pulsed simultaneously every two weeks (NI), and nutrients and insecticide pulsed in alternating weeks (N_I). Insecticide and nutrient pulses consisted of 2 μg/L of chlorpyrifos, 560 μg/L of nitrogen, and 39.9 μg/L of phosphorus. Zooplankton abundance, community structure, and diversity were used as structural indicators. Chlorophyll a and net production were used as functional indicators. We found no effect of the treatments on zooplankton abundance, while richness and Shannon diversity was lower in treatments with pulsed insecticide (NI and N_I) compared to control treatment (N). Phytoplankton biomass was higher in the treatments with insecticide than in the controls (N). Higher phytoplankton biomass could be explained by an indirect effect shift from a cladocera-dominated to a copepod-dominated community in response to the insecticide treatment. Overall, the insecticide disturbance had direct and indirect effects on the community and did not depend on whether insecticides were pulsed synchronously or asynchronously with nutrients.

Remote mapping of leafy spurge (Euphorbia esula, L.) in Northwestern Colorado

Leafy spurge (Euphorbia esula L.) has been introduced to the Yampa River in Northwestern Colorado for over 40 years and flood and runoff events transport leafy spurge propagules onto adjacent landscapes. The spread of leafy spurge beyond the river channels has yet to be mapped and recorded, and this research was conducted to map leafy spurge occurrence in the Yampa River Valley. Significant stakeholder mapping efforts took place in the summer of 2019–2021, leading to excellent spatial data on leafy spurge presence and absence along the main channel. In summer 2019, multispectral SPOT seven satellite imagery, stakeholder ground mapping efforts, and bright yellow-green leafy spurge bracts were used to interpret imagery, identify dense, unobscured patches of leafy spurge, and digitize them. Spectral signatures from training samples for leafy spurge and other land cover classes (generalized as “not leafy spurge”) were then used to train a Random Forest machine learning classification. In the summer of 2021, generated classification maps were compared to multispectral satellite imagery and stakeholder ground mapped leafy spurge presence. Mismatches were identified, and 271 validation locations were identified, navigated to, and evaluated for leafy spurge presence. Leafy spurge training samples were classified with 96% accuracy. Correctly classified leafy spurge locations had higher leafy spurge coverage and lower overstory canopy than missed leafy spurge locations. Leafy spurge growing beneath shrub canopy or growing as individual plants along the riverbanks were more likely to be missed. A frequency analysis for other plant species found at validation locations determined that smooth brome (Bromus inermis Leyss.), dandelion (Taraxacum officinale L.), and willow (Salix sp.) were most frequently misclassified as leafy spurge. In conclusion, multispectral satellite imagery was useful at remote detection of leafy spurge in open areas with dense leafy spurge coverage, but more work must be done for identification of sparse and diffuse leafy spurge infestations.

Spatiotemporal variations and overflow risk analysis of the Salt Lake in the Hoh Xil Region using machine learning methods

Global warming is inducing dramatic changes in fluvial geomorphology and reshaping the hydrological connections between rivers and lakes. The water level and area of the Salt Lake have increased rapidly since the outburst of the Zonag Lake in the Hoh Xil region of the Qinghai–Tibet Plateau in 2011, threatening the downstream infrastructure. However, fewer studies have focused on its spatiotemporal variation and overflow risk over long time series. Here, we used three machine learning algorithms: Classification and Regression Trees (CART), Random Forest (RF), and Support Vector Machine (SVM) to extract the area of the Salt Lake for a long time series, analyzed its spatiotemporal variation from 1973 to 2021, and finally assessed the overflow risk. The Kappa coefficient (KAPPA) and the overall accuracy (OA) were used to evaluate the performance of the models. The results showed that Random Forest performs superior in lake extraction (KAPPA = 0.98, overall accuracy = 0.99), followed by Classification and Regression Trees and Support Vector Machine. normalized difference water index is the relatively important feature variable in both RF and CART. Before the outburst event, the area change of the Salt Lake was consistent with the variation in precipitation; after that, it showed a remarkable area increase (circa 350%) in all orientations, and the main direction was the southeast. Without the construction of the emergency drainage channel, the simulation result indicated that the earliest and latest times of the Salt Lake overflow event are predicted to occur in 2020 and 2031, respectively. The results of this paper not only demonstrate that RF is more suitable for water extraction and help understand the water system reorganization event.

Pollutant characteristics and size distribution of trace elements during stormwater runoff events

Priority trace elements As, Cu, Cr, Ni, Pb, and Zn are largely present in urban stormwaters and should be controlled due to their impact on water bodies. As it has been observed that there are numerous limitations in the purification performance of roadway stormwater runoff treatment systems, the particulate size distribution of trace elements was studied extensively to assess the colloidal fractions present in stormwater runoff of a high-traffic roadway site (100 000 vehicles/day). Besides the evaluation of the total Event Mean Concentrations (EMCs) and dissolved concentrations, a complete fractionation scheme of low settling phase of runoff water (<30 µm) was applied to a series of rain events, considering 8 µm and 5 kDa, as relevant cut-offs between particulate/colloidal elements, and colloidal/truly dissolved elements. The intra-event size distribution was also considered to assess the first-flush effect and the variation of the physical speciation. The high-level traffic associated with climatic conditions (rainfall depth) are responsible for very high EMCs for Zn and Cu, together with a significant exceedance of the European environmental water standards (<0.45 µm concentrations). Ni, Cr and Pb are mainly present as particulate and colloidal metals. Therefore, considering a more complete size distribution is relevant for these elements. For As, Cu and Zn, the conventional fractionation at 0.45 µm seems to be sufficient to evaluate the highly mobile fraction. The results also indicate that, while the main particulate load may be collected within the early volumes of the rainfall events (about 30%), the variability of rain intensity during long events induces changes in particulate and colloidal metals in the runoff. The treatment of runoff must then be considered as a whole in view of the variability of trace metal size distribution. This suggests that urban stormwater runoff treatment systems which employ only the settleable fraction are not sufficient to limit the chemical and ecological risks to receiving water bodies.

Contaminant Exposure and Transport from Three Potential Reuse Waters within a Single Watershed.

Global demand for safe and sustainable water supplies necessitates a better understanding of contaminant exposures in potential reuse waters. In this study, we compared exposures and load contributions to surface water from the discharge of three reuse waters (wastewater effluent, urban stormwater, and agricultural runoff). Results document substantial and varying organic-chemical contribution to surface water from effluent discharges (e.g., disinfection byproducts [DBP], prescription pharmaceuticals, industrial/household chemicals), urban stormwater (e.g., polycyclic aromatic hydrocarbons, pesticides, nonprescription pharmaceuticals), and agricultural runoff (e.g., pesticides). Excluding DBPs, episodic storm-event organic concentrations and loads from urban stormwater were comparable to and often exceeded those of daily wastewater-effluent discharges. We also assessed if wastewater-effluent irrigation to corn resulted in measurable effects on organic-chemical concentrations in rain-induced agricultural runoff and harvested feedstock. Overall, the target-organic load of 491 g from wastewater-effluent irrigation to the study corn field during the 2019 growing season did not produce substantial dissolved organic-contaminant contributions in subsequent rain-induced runoff events. Out of the 140 detected organics in source wastewater-effluent irrigation, only imidacloprid and estrone had concentrations that resulted in observable differences between rain-induced agricultural runoff from the effluent-irrigated and nonirrigated corn fields. Analyses of pharmaceuticals and per-/polyfluoroalkyl substances in at-harvest corn-plant samples detected two prescription antibiotics, norfloxacin and ciprofloxacin, at concentrations of 36 and 70 ng/g, respectively, in effluent-irrigated corn-plant samples; no contaminants were detected in noneffluent irrigated corn-plant samples.

Hydrological Processes in a Small Research Watershed under Forest Coverage in the Coast of Chiapas, Mexico

In the hydrological watershed, some natural processes take place in which the interaction of water, soil, climate and vegetation favors the capture of water. The present study aimed to evaluate preliminary information regarding the hydrological response and the water balance in a small research watershed with tropical forest cover (15°01'44''N and 92°13'55''W, 471 m, 2.3 has). Events of precipitation, direct runoff, infiltration rate and baseflow were performed. The amount, duration and intensity of rainfall events were recorded with the use of a pluviograph. Surface runoff was quantified with an established gauging station, an H-type gauging device and a horizontal mechanical gauging limnograph. Runoff base flow was measured at the gauging station using the volume-time method. Infiltration was measured using a triple ring infiltrometer, taking two measurements in the upper part and two in the lower part of the microbasin. Evapotranspiration was measured with the amount of rainfall entering and runoff leaving the watershed. In the study period, annual rainfall of 4417.6 mm distributed over 181 events were recorded; about 70% of the storms showed lower intensities at 20 mm·h-1. The total runoff was 345.8 mm caused by half of the rainfall events, which represents 7.8% of the total rain; 77% of runoff events showed lower sheets of 5 mm and an average specific rate of 20.7 L·s-1·ha-1 with a maximum of 113.6 L·s-1·ha-1. Three runoff events were greater than 20.1 mm and caused the 22.5% of the total runoff depth in the study period showing the equilibrium conditions in the hydrological response of the forest. Water outputs like baseflow was 669.5 mm. In this way, 90% of the rainfall is infiltrated every year in the micro-watershed, which shows the importance of the plant cover in the hydrological regulation and the groundwater recharge.

Measuring and Modeling Impacts of Gravel Road Design on Sediment Generation in the Southeastern U.S.

Highlights The erodibility of heavily trafficked gravel roads can be much greater than that of low volume forest roads. Improved designs of heavily trafficked gravel roads can decrease sediment generation by more than 90 percent. The WEPP Model can be successfully parameterized for high traffic gravel roads to reflect the effects of weather, road design, and topography. Abstract. The purposes of this study were to support a watershed modeling analysis by evaluating the ability to the Water Erosion Prediction Project (WEPP) model to estimate sediment generated by high traffic gravel roads, and to determine the erodibility of two designs of high-traffic gravel roads. In many watersheds, the road network can be a major source of sediment. The ability to predict erosion from roads, evaluate the effects of design and management on road sedimentation, and compare sediment from roads to other sources of sediment in the watershed is an ongoing need by watershed managers. The Water Erosion Prediction Project (WEPP) model is a widely used model for predicting sediment from forest roads. There has, however, been little information published on erosion from high traffic gravel roads and WEPP applications to such roads. To evaluate road erosion predictions, a study was conducted incorporating two road designs at Fort Benning, Georgia, U.S. One design followed a common practice of starting with a native material road and adding gravel and grading as required. Erosion and rutting on the road surface were common occurrences on this type of road. The improved design was a “graded aggregate base” design, built with compacted aggregate layers. To evaluate erosion risks for these two road designs, runoff and sediment delivery were measured from ten plots ranging in size from 63 to 150 m2. Runoff depths up to 50 mm occurred from daily rainfall amounts up to nearly 60 mm, with least square mean event runoff values of 6.5 mm from unimproved plots and 14.9 mm from improved road plots. Delivered sediment ranged from zero to 18 Mg ha-1 from individual storms with least square mean amounts of 2.27 Mg ha-1 of sediment delivered from unimproved road plots compared to only 0.026 Mg ha-1 delivered from improved road design plots for a given runoff event. Hydraulic conductivity was found by calibration to be 3.0 mm h-1 for unimproved roads and 1.3 mm h-1 for improved road segments. Rill erodibility was 0.09 s m-1 for unimproved roads and 0.0008 s m-1 for improved roads, values that were greater than had been measured on road erosion studies elsewhere that were typically less than 0.0004 s m-1. The critical shear for the unimproved roads was the minimum that the WEPP model would accept, 0.0001 Pa, but was a more typical value of 1.5 Pa for the improved road segments. When applying the calibrated erodibility values to a validation data set, the Willmott indices of agreement were 0.62 and 0.82 for runoff for unimproved and improved roads, respectively, and 0.67 and 0.66 for sediment delivery from unimproved and improved roads, respectively, indicating good agreement between observed and WEPP-estimated runoff and erosion rates. A sensitivity analysis and calibration analysis found that the WEPP model was not sensitive to interrill erosion for this application. A sensitivity analysis coupled with a WEPP validation analysis showed that WEPP could incorporate weather, topography, soil, and road design features to predict sediment delivery from highly erodible road segments. The study suggests that there is a need for a simulated runoff study to determine high values of rill erodibility more precisely on unimproved high traffic roads, and that there is a need to incorporate more erodible road erodibility values into the online WEPP:Road interface for the WEPP model. The road erosion rates and effectiveness of improved road designs for reducing off-road sediment reported in this study will be useful to managers seeking to quantify and reduce road erosion rates from high-traffic gravel roads in sensitive watersheds. Keywords: Erodibility, Gravel Roads, Soil Erosion, WEPP.

Effectiveness of Flue Gas Desulfurization Gypsum Reused as a Litter Amendment in Broiler House to Mitigate Phosphorus Loss During Runoff Events

Effectiveness of Flue Gas Desulfurization Gypsum Reused as a Litter Amendment in Broiler House to Mitigate Phosphorus Loss During Runoff Events

A robust method to identify the occurrence of a runoff-generated debris flow

Debris flows generated by rainfall runoff can occur in rocky alpine landscapes and burned steeplands. Runoff-generated debris-flow events are commonly composed of a series of dense granular surge fronts separated by water-rich flows. Owing to this intra-event variability in flow composition and mechanics, post-event interpretations of preserved sedimentary deposits, or lack thereof, can result in a dizzying mix of interpretations that range from clearwater flow to debris flow. Accurate identification of the presence or absence of a debris flow during a runoff event is critical for building empirical models used to predict likelihood of debris-flow occurrence, rainfall thresholds, and flow properties. Here, we propose a simple, quantitative method to identify the occurrence of a runoff-generated debris flow, based on a dimensionless discharge Q* calculated as the ratio of the peak event discharge Qp to the theoretical maximum clearwater runoff rate Qw. Using a preliminary compilation of Q* values from floods and runoff-generated debris flows, we find 98% of floods have Q* values &lt; 1.6, whereas 91% of debris flows have Q* values greater than 1.6. Estimating Q* is typically straightforward as part of standard post-event reconnaissance if suitable rainfall estimates are available, and appears to be a robust indicator that runoff-generated debris flows traversed a particular portion of a valley network.

Monitoring Blockage and Overflow Events in Small-Sized Sewer Network Using Contactless Flow Sensors in Hong Kong: Problems, Causes, and Proposed Solution

Monitoring Blockage and Overflow Events in Small-Sized Sewer Network Using Contactless Flow Sensors in Hong Kong: Problems, Causes, and Proposed Solution

Analysis of Surrogate Physicochemical Parameters for Studying Heavy Metal Pollution in Urban Road Runoff

The pollution associated with road runoff water can generate significant impacts on the receiving natural environment due to the significant masses mobilized under certain climate, morphological, and anthropic conditions. The aim of this paper is to show an analysis of the possible surrogate conventional physicochemical parameters of pollution by heavy metals (HMs) in urban road runoff. The best surrogate physicochemical parameters are detected by a differentiated analysis between the HM concentrations (Fe, Al, As, Ba, Cd, Co, Cu, Cr, Mn, Hg, Ni, Pb, V, and Zn) in the total, particulate, and dissolved fractions. This analysis is also performed under two scenarios of runoff event energy according to the mobilized TSS load. The results suggested that it was easier to detect surrogate parameters for total HM concentrations during higher-energy runoff events. The outcomes hinted that regardless of the runoff event energy, it was easier to detect conventional surrogate parameters for the particulate HM concentration compared to the dissolved HM concentration. The findings showed for total HM concentration that the best surrogate parameter during higher-energy runoff events was TSS. The best surrogate HM during these runoff events was Fe. The results also suggested that HMs with high percentages of association with the particulate fraction (&gt;70%) of road runoff were the best surrogates for the other HMs under study. For lower-energy runoff events, the best surrogate parameter was VSS, although TSS also showed good behavior.

Disentangling scatter in long-term concentration–discharge relationships: the role of event types

Abstract. Relationships between nitrate concentrations and discharge rates (C–Q) at the catchment outlet can provide insights into sources, mobilization and biogeochemical transformations of nitrate within the catchment. Nitrate C–Q relationships often exhibit considerable scatter that might be related to variable hydrologic conditions during runoff events at sampling time, corresponding to variable sources and flow paths despite similar discharge (Q) rates. Although previous studies investigated the origins of this scatter in individual or in a few catchments, the role of different runoff event types across a large set of catchments is not yet fully understood. This study combines a hydrological runoff event classification framework with low-frequency nitrate samples in 184 catchments to explore the role of different runoff events in shaping long-term C–Q relationships and their variability across contrasting catchments. In most of the catchments, snow-impacted events produce positive deviations of concentrations, indicating an increased nitrate mobilization compared to the long-term pattern. In contrast, negative deviations occur mostly for rainfall-induced events with dry antecedent conditions, indicating the occurrence of lower nitrate concentrations (C) in river flows than their long-term pattern values during this type of event. Pronounced differences in event runoff coefficients among different event types indicate their contrasting levels of hydrologic connectivity that in turn might play a key role in controlling nitrate transport due to the activation of faster flow paths between sources and streams. Using long-term, low-frequency nitrate data, we demonstrate that runoff event types shape observed scatter in long-term C–Q relationships according to their level of hydrologic connectivity. In addition, we hypothesize that the level of biogeochemical attenuation of catchments can partially explain the spatial variability of the scatter during different event types.

Hydroclimatic variability across the international Lake of the Woods watershed: Implications for nutrient export and climate sensitivity

Characterizing streamflow and relationships with climate and watershed characteristics is an essential first step in the design of any monitoring program to assess basin response to changes in land use or climate. This is especially true for the international Lake of the Woods watershed, where recurrent algae blooms have been associated with nutrient inputs from the watershed and climate warming. Here, we present a basin-wide hydroclimatic analysis within the sparsely monitored Canadian portion of the basin. Spatial and temporal patterns in climate and runoff were assessed across the two major geo-zones: the Precambrian ‘Shield zone’, dominated by bedrock, forests and lakes, and the poorly drained ‘Agassiz zone’ where ditching and drainage for agriculture have substantially enhanced the hydrologic connectivity. While climate conditions were consistent across the watershed, Agassiz basins were flashy, highly variable, and more seasonal compared with Shield rivers, likely due to the moderating effect of lake storage in the Shield region. Temperatures increased across the basin (1910–2010), and there was more rainfall and runoff during the ice-covered months (Nov-Mar), suggesting a shift toward earlier snowmelt. Marked seasonality and large swings in flow extremes at the Agassiz rivers suggest this region is particularly sensitive to hydroclimatic change and that frequent monitoring is needed to capture important periods of nutrient export like spring runoff and storm events. In contrast, substantial storage within the Shield landscape suggests this zone is more hydrologically ‘resilient’ to climate extremes and that water quality and quantity measurements can be less frequent.

Seasonal Fluctuations in Nitrate Levels Can Trigger Lead Solder Corrosion Problems in Drinking Water.

After a utility switched its source water from ground to surface water in 2017, first draw water lead levels spiked due to increased lead solder corrosion that could not be explained by existing knowledge. When lead release was not adequately reduced with a 90:10 orthophosphate/polyphosphate corrosion inhibitor blend or even high levels of 100% orthophosphate, an in-depth investigation of possible causes revealed a strong correlation between 90th percentile lead and seasonal fluctuations in surface water nitrate levels. Complementary bench-scale studies that tested new copper coupons with lead solder and harvested pipes from a worst case home verified a strong relationship between nitrate and elevated lead. Lead release in the presence of nitrate became increasingly erratic with time, resulting in the spalling of large lead solder particulates up to 7 mm in length into the water. Lead levels were occasionally >1000 ppb in homes and >100000 ppb in the bench experiments with harvested pipe. Orthophosphate was unable to sufficiently reduce lead levels below the action level during periods with high nitrate levels in the bench studies. Water utilities and regulators should proactively consider possible unintended consequences of higher nitrate levels on lead release when changing source waters or during seasonal runoff events.

Passive Treatment of Circumneutral Mine Drainage from the St. Louis Mine Tunnel, Rico, CO: Part 1—Case Study: Characteristics of the Mine Drainage

AbstractThis publication is a case study of the seasonal variability of mine water drainage from the Saint Louis Tunnel (SLT) at the inactive Rico-Argentine mine site located in southwestern Colorado. It is an introductory paper for the two passive water treatment system technology evaluations contained in this issue. Mine water chemistry changes from baseflow to a snowmelt runoff event (SMRE) where snowmelt runoff follows preferential migration pathways to flush acidic weathering products from the upper mine workings to the SLT. Baseflow mine drainage is characterized as circumneutral, with Zn, Cd, Mn, and Ni concentrations primarily in the dissolved form. Dissolved Zn, Mn, Fe, and potentially Cd illustrate equilibrium with carbonate minerals. Total concentrations of Fe, Cu, Pb, and As are primarily in the suspended form and suggest sorption to Fe oxides. Mine water chemistry during the SMRE reflects mixing of circumneutral baseflow waters with more acidic waters flushing the upper mine workings. Geothermal activity provides for a consistently warm mine water discharge from the SLT. The two seasons that provide the most challenge to passive water treatment of SLT mine drainage are the SMRE period and the low flow stage of the Dolores River. Mine water flow and chemistry during SMRE are highly correlated with Dolores River flow and this site conceptual model was and will be used to assist in pilot project evaluation, water treatment system design, monitoring system design, a seasonal compliance approach, and water management.

Optimizing phosphate fertilizer input to reduce phosphorus loss in rice-oilseed rape rotation.

Identifying the major sources and critical periods of P loss from agricultural fields provides important guidance for reducing P loss. A rice-oilseed rape rotation with no P fertilization (NP, control), medium P fertilization (MP, 90kg P2O5 ha-1 season-1), and high P fertilization (HP, 180kg P2O5 ha-1 season-1) was conducted from 2019 to 2021 in the middle Yangtze River Basin. Runoff and leaching P losses were measured simultaneously using runoff event monitoring and a percolation device. Applying P fertilizer increased the P concentration in the field ponding water and percolation water of the rice-oilseed rape rotation. During the rice growing season, total P (TP), dissolved P (DP), and particulate P (PP) concentrations in the field ponding water and percolation water peaked 1day after P was applied, and then decreased rapidly. After 10days of fertilization, P concentration in the field ponding water of the MP treatment decreased to a minimum and stabilized, while the HP treatment extended this period to 20days. The highest P concentration in percolation water was observed at the first sampling during the oilseed rape season, and then it continued to decrease. Inputting P fertilizer increased P loss by 55.0-109.9% compared to the NP treatment, with annual P losses of 0.89-1.10kg P ha-1, of which runoff loss accounted for 61.7-62.9%. Fertilization and precipitation resulted in varied P loss within and between seasons. Runoff from heavy precipitation during the rice season was the main source of P loss, while PP accounted for 54.7-77.6% of runoff P loss. The strong utilization of soil P by rice resulted in a lower demand for exogenous P fertilizer than oilseed rape. Excessive P input increased the soil P surplus and vertical migration. Therefore, reducing rice season P fertilizer inputs to achieve annual P balance in rice-oilseed rape rotation can effectively reduce soil P surplus and loss while ensuring crop P demand, and the initial 10 d after fertilization in the rice season was a critical period for reducing P runoff loss.

Effects of watershed-scale green infrastructure retrofits on urban stormwater quality: A paired watershed study to quantify nutrient and sediment removal

Urban stormwater represents a substantial source of nutrients and sediment to aquatic ecosystems. Green infrastructure (GI), including bioretention and permeable pavement, is an increasingly utilized method to treat stormwater pollutants. Using soil and plants as natural filters, these systems are effective at the site scale, but little evidence exists regarding their performance at the watershed-scale. Blueprint Columbus is an effort by the City of Columbus, Ohio, USA to retrofit GI to eliminate sanitary sewer overflows and remove 20% of total suspended solids (TSS) from existing developed areas. Changes in water quality resulting from the combined effects of many GI practices installed in 11.5 and 47.8 ha treatment watersheds were quantified using a paired-watershed approach. Based on water quality data collected by automated samplers over a 3.5-year period, significant reductions in particulate and dissolved nutrients as well as sediment were observed following the installation of GI in both treatment watersheds compared to the control. Total nitrogen (TN), phosphorus (TP), and TSS concentrations decreased by 13.7–24.1%, 20.9–47.4%, and 61.6–67.7%, respectively. Runoff attenuation by GI contributed to pollutant load reductions of 24.0–25.4% (TN), 27.8–32.6% (TP), and 59.5–78.3% (TSS). Orthophosphate concentrations and loads increased in the watershed with bioretention only but significantly decreased in the treatment watershed with bioretention and permeable pavement. Reductions in TSS concentration were similar (within a margin of 5%) to the percent of the watershed imperviousness treated by GI. Results demonstrate that GI was effective in reducing runoff event mean concentrations and loads at the watershed-scale.

Hybrid model for daily runoff interval predictions based on Bayesian inference

The highly nonlinear and nonstationary nature of runoff events in changing environments makes accurate and reliable runoff forecasting difficult. We propose a hybrid model by integrating an autoregressive model, Bayesian inference, a complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) algorithm, Bayesian optimization, and support vector regression. Two Bayesian inference methods (the No-U-Turn Sampler (NUTS) and variational inference) were used to calculate the parameters of the AR model to obtain a Bayesian AR (BAR) model. Credible intervals were used to analyze the uncertainty of the parameters and model prediction results. The above model is applied to the daily runoff predictions of hydrological stations in the Yellow River Basin of China. The results show that (1) the hybrid model can improve the prediction accuracy and (2) the NUTS algorithm-based model provides a narrower reliable interval and performs better in uncertainty analyses.