Random forest-based modeling of stream nutrients at national level in a data-scarce region

The Science of The Total Environment, Journal Year: 2022, Volume and Issue: 840, P. 156613 - 156613

Published: June 11, 2022

Nutrient runoff from agricultural production is one of the main causes water quality deterioration in river systems and coastal waters. Water modeling can be used for gaining insight into issues order to implement effective mitigation efforts. Process-based nutrient models are very complex, requiring a lot input parameters computationally expensive calibration. Recently, ML approaches have shown achieve an accuracy comparable process-based even outperform them when describing nonlinear relationships. We observations 242 Estonian catchments, amounting 469 yearly TN 470 TP measurements covering period 2016-2020 train random forest (RF) predicting annual N P concentrations. total 82 predictor variables, including land cover, soil, climate topography applied feature selection strategy reduce number dependent features models. The SHAP method was deriving most relevant predictors. performance our previous Baltic region with model having R2 score 0.83 0.52, respectively. However, as data easier obtain, offer superior applicability areas, where availability insufficient approaches. Therefore, enable give robust estimation losses at national level allows capture spatial variability which turn enables provide decision-making support regional management plans.

Language: Английский

---

Catchment concentration–discharge relationships across temporal scales: A review

Wiley Interdisciplinary Reviews Water, Journal Year: 2023, Volume and Issue: 11(2)

Published: Dec. 7, 2023

Abstract Processes that drive variability in catchment solute sourcing, transformation, and transport can be investigated using concentration–discharge (C–Q) relationships. These relationships reflect in‐stream processes operating across nested temporal scales, incorporating both short long‐term patterns. Scientists therefore leverage catchment‐scale C–Q datasets to identify distinguish among the underlying meteorological, biological, geological export patterns from catchments influence shape of their respective We have synthesized current knowledge regarding geological, meteorological on for various types diel decadal time scales. cross‐scale linkages tools researchers use explore these interactions Finally, we gaps our understanding dynamics as reflections processes. also lay foundation developing an integrated approach investigate relationships, reflecting biogeochemical effects environmental change water quality. This article is categorized under: Science Water > Hydrological Quality Environmental Change

Language: Английский

---

High-intensity rainfall following drought triggers extreme nutrient concentrations in a small agricultural catchment

Water Research, Journal Year: 2024, Volume and Issue: 264, P. 122108 - 122108

Published: July 25, 2024

The profound influence of climate change on the hydrological cycle raises concerns about its potential impacts water quality, particularly in agricultural catchments. Here, we analysed 200 storm events monitored for nitrate and total phosphorus (TP) at sub-hourly intervals from 2016 to 2023 Kervidy-Naizin catchment (north-western France). Using Extreme Value theory, identified with extreme concentrations compared their hydroclimatic characteristics those non-extreme events. We hypothesised that concentration occurred under conditions, which are projected become more frequent future. showed dilution patterns nitrate, decreasing by up 41 %, accretion TP, increasing 1400 % Hydroclimatic conditions during were characterised high rainfall intensities low antecedent discharge, but no particular mean discharge. During events, concentration-discharge relationships exhibited primarily clockwise hysteresis, whereas TP displayed an equal mix anticlockwise loops. In contrast, hysteresis weak TP. interpreted these dynamics controls as result infiltration-excess overland flow diluting nitrate-rich groundwater exporting large amounts intensive following droughts, while fluctuations riparian zone streambed remobilization control nutrient exports Given frequency intensity extremes, such retrospective analyses can provide valuable insights into future streams draining

Language: Английский

---

QUADICA: water QUAlity, DIscharge and Catchment Attributes for large-sample studies in Germany

Earth system science data, Journal Year: 2022, Volume and Issue: 14(8), P. 3715 - 3741

Published: Aug. 17, 2022

Abstract. Environmental data are the key to defining and addressing water quality quantity challenges at catchment scale. Here, we present first large-sample set for 1386 German catchments covering a large range of hydroclimatic, topographic, geologic, land use, anthropogenic settings. QUADICA (water QUAlity, DIscharge Catchment Attributes studies in Germany) combines with data, meteorological nutrient forcing attributes. The comprises time series riverine macronutrient concentrations (species nitrogen, phosphorus, organic carbon) diffuse nitrogen (nitrogen surplus, atmospheric deposition, fixation) Time generally aggregated an annual basis; however, 140 stations long-term (more than 20 years), additionally monthly median discharge concentrations, flow-normalized corresponding mean fluxes as outputs from Weighted Regressions on Time, Discharge, Season (WRTDS). attributes include inputs point sources characteristics topography, climate, cover, lithology, soils. This comprehensive, freely available collection spatial temporal coverage can facilitate data-driven assessments scale well mechanistic modeling studies. is https://doi.org/10.4211/hs.0ec5f43e43c349ff818a8d57699c0fe1 (Ebeling et al., 2022b) https://doi.org/10.4211/hs.88254bd930d1466c85992a7dea6947a4 2022a).

Language: Английский

---

Soil CO₂ Controls Short‐Term Variation but Climate Regulates Long‐Term Mean of Riverine Inorganic Carbon

Global Biogeochemical Cycles, Journal Year: 2022, Volume and Issue: 36(8)

Published: Aug. 1, 2022

Abstract The evasion of CO 2 from inland waters, a major carbon source to the atmosphere, depends on dissolved inorganic (DIC) concentrations. Our understanding DIC dynamics across gradients climate, geology, and vegetation conditions however have remained elusive. To understand its large‐scale patterns drivers, we collated instantaneous mean (multiyear average) concentrations about 100 rivers draining minimally‐impacted watersheds in contiguous United States. Within individual sites, ( C ) measured at daily seasonal scales exhibit near‐universal response changes river discharge Q negative power law form. High occur low when DIC‐enriched groundwater dominates discharge; under high flow relatively DIC‐poor shallow soil water predominates discharge. Such echo widely observed increase with depth shallow‐and‐deep hypothesis that emphasizes importance paths chemistry. Across m decrease increasing ), long‐term climate measure, reachs maxima around 200 mm/yr. A parsimonious model reveals arises accumulation rates DIC‐generating reactions are compared export fluxes arid climates. Although similarly discharge, results here highlight their distinct drivers: seasonal‐scale concentration variations (C) controlled by subsurface distribution over (from below), whereas regulated above). emphasize significance land‐river connectivity via paths. They also underscore characterizing illuminate belowground processes order project future cycles warming climate.

Language: Английский

---

When good signatures go bad: Applying hydrologic signatures in large sample studies

Hydrological Processes, Journal Year: 2023, Volume and Issue: 37(9)

Published: Sept. 1, 2023

Abstract Hydrologic signatures are quantitative metrics that describe streamflow statistics and dynamics. Signatures have many applications, including assessing habitat suitability hydrologic alteration, calibrating evaluating models, defining similarity between watersheds investigating watershed processes. Increasingly, being used in large sample studies to guide flow management modelling at continental scales. Using involving 1000s of brings new challenges as it becomes impractical examine signature parameters behaviour each watershed. For example, we might wish check describing flood event characteristics correctly identified periods, values not been biassed by data errors, or human natural influences on interpreted. In this commentary, draw from our collective experience present case where naïve application fails identify These include unusual precipitation regimes, quality issues, use human‐influenced watersheds. We conclude providing guidance recommendations applying studies.

Language: Английский

---

Winter post-droughts amplify extreme nitrate concentrations in German rivers

Environmental Research Letters, Journal Year: 2024, Volume and Issue: 19(2), P. 024007 - 024007

Published: Jan. 2, 2024

Abstract Hydrological extremes can affect nutrient export from catchments to streams, posing a threat aquatic ecosystems. In this study, we investigated the effects of hydrological drought on nitrate concentrations in streamflow 182 German 1980 2020. We found that across all seasons, 40% and 25% showed significantly lower during post-droughts, respectively, when compared non-drought conditions. However, observed pronounced spatial variability responses, particularly winter droughts with more exhibiting higher concentrations. Specifically, were study droughts, wetter low nitrogen retention. During are 19% catchments, especially surplus. Moreover, likelihood seasonal increased by 6% post-drought our catchments. Considering projected increase frequency Germany, corresponding periods poses potential ecosystem health.

Language: Английский

---

Nitrate Transport and Retention in Western European Catchments Are Shaped by Hydroclimate and Subsurface Properties

Water Resources Research, Journal Year: 2021, Volume and Issue: 57(10)

Published: Sept. 28, 2021

Abstract Excess nitrogen (N) from anthropogenic sources deteriorates freshwater resources. Actions taken to reduce N inputs the biosphere often show limited or only delayed effects in receiving surface waters hinting at large legacy stores built up catchments' soils and groundwater. Here, we quantify transport retention of 238 Western European catchments by analyzing a unique data set long‐term input output time series. We find that half exhibited times with an average riverine concentration peak arriving 5 years after application. Longer were evident high potential evapotranspiration low precipitation seasonality. On retained 72% (Interquartile Range: 18%) diffuse efficiency being specifically discharge thick, unconsolidated aquifers. The estimated scales do not explain observed retention, suggesting dominant role biogeochemical rather than store Future water quality management should account for accumulated fostering denitrifying conditions soil recycling avoid leaching deteriorations decades come.

Language: Английский

---

Disparate Seasonal Nitrate Export From Nested Heterogeneous Subcatchments Revealed With StorAge Selection Functions

Water Resources Research, Journal Year: 2022, Volume and Issue: 58(3)

Published: March 1, 2022

Abstract Understanding catchment controls on solute export is a prerequisite for water quality management. StorAge Selection (SAS) functions encapsulate essential information about functioning in terms of discharge selection preference and dynamics. However, they lack the spatial origin solutes when applied at scale, thereby limiting our understanding internal (subcatchment) functioning. Here, we parameterized SAS spatially explicit way to understand responses transport dynamics reactive dissolved nitrate (N‐NO 3 ). The model was nested mesoscale (457 km 2 ), consisting mountainous partly forested, agricultural subcatchment, middle‐reach forested lowland subcatchment. captured flow concentration not only outlet but also gauging stations. Results reveal disparate subsurface mixing among headwater subcatchments. subcatchment has high seasonal variation schemes younger discharge, while less pronounced seasonality much older discharge. Consequently, from shows strong seasonality, whereas that stable time. temporally varying subcatchments alternate dominant contribution low‐flow periods between Overall, results demonstrate modeling provides useful functioning, helping develop or evaluate management practices.

Language: Английский

---

Effects of 66 years of water management and hydroclimatic change on the urban hydrology and water quality of the Panke catchment, Berlin, Germany

The Science of The Total Environment, Journal Year: 2023, Volume and Issue: 900, P. 165764 - 165764

Published: July 27, 2023

Language: Английский

---