Project

Back to overview

HYPOTHESIS TESTING USING CONTROLLED EXPERIMENTS TO CHARACTERIZE DIFFUSE POLLUTION IN SMALL AGRICULTURAL CATCHMENTS

Applicant Fenicia Fabrizio
Number 163322
Funding scheme Project funding (Div. I-III)
Research institution Eawag
Institution of higher education Swiss Federal Institute of Aquatic Science and Technology - EAWAG
Main discipline Hydrology, Limnology, Glaciology
Start/End 01.04.2016 - 31.03.2020
Approved amount 236'554.00
Show all

All Disciplines (2)

Discipline
Hydrology, Limnology, Glaciology
Agricultural Engineering

Keywords (7)

fieldwork; herbicides; hypothesis testing; flexible frameworks; hydrological modelling; water quality modelling; diffuse pollution

Lay Summary (Italian)

Lead
Gli erbicidi sono sostanze chimiche usate nell’agricoltura moderna per ridurre lo sviluppo di vegetazione indesiderata. Queste sostanze, attraverso vari percorsi, possono raggiungere aque sotterranee, fiumi e laghi. La presenza degli erbicidi in queste acque puo‘ condurre a disfunzioni dell’ecosistema, perdita di biodiversita‘, e problemi alla salute dell’uomo. Una migliore comprenzione dei fattori che controllano il trasporto degli erbicidi nell’ambiente naturale e‘ fondamentale allo sviluppo di efficienti misure di mitigazione.
Lay summary

Contenuti e obbiettivi

In questo progetto, combiniamo nuovi approcci di modellazione e di stima delle incertezze per modellare il trasporto degli erbicidi in due piccoli bacini nel Plateau della Svizzera. Questi bacini sono stati equipaggiati per condurre comprensivi esperimenti sul trasporto degli erbicidi. Gli obbiettivi di questo progetto sono: (1) raggiungere una migliore comprensione di come i dati sperimentali aiutino il processo di modellazione, (2) approfondire la comprensione dei principali fattori che controllano il trasporto degli erbicidi, attraverso lo sviluppo di vari modelli matematici, (3) determinare i livelli di incertezza associati alla previsione della concentrazione degli erbicidi nei fiumi.

Contesto scientifico e sociale

La presenza di erbicidi in acque superficiali e sotterranee e’ un importante problema ambientale. La efficacia di misure di mitigazione e’ strettamente legata al livello di comprensione dei processi idrologici e chimici nell’ambiente. Le analisi condotte in questo progetto sono intese a migliorare la conoscenza dei processi legati al trasporto degli erbicidi nell’ambiente natural.

Parole chiave

Modellazione idrologica, modellazione di qualita’ dell’acqua, erbicidi, inquinamento diffuso, test di ipotesi, modelli flessibili, esperimenti sul campo.

Direct link to Lay Summary Last update: 01.02.2016

Lay Summary (English)

Lead
Herbicides are chemicals substances used in modern agriculture to reduce the development of unwanted vegetation. These substances, through various flow pathways, may reach underground waters, streams, and lakes. The presence of herbicides in these water bodies may lead to ecosystem dysfunction, loss of biodiversity, and human water security threats. A better understanding of the factors that control herbicides transport in the natural environment is critical to the development of efficient mitigation measures.
Lay summary

Content and objectives

In this project, we will combine multi-model approaches and novel uncertainty estimation techniques to model herbicide transport in two headwater catchments in the Swiss plateau. These catchments have been the location of unique controlled herbicide experiments. The expected outcomes of this project consist of (1) a better understanding of the contribution of experimental data in informing the modelling process of diffuse pollution from small agricultural catchments, (2) an improved understanding of the dominant controls on herbicide transport and of appropriate conceptualizations of catchment scale processes and (3) an assessment of the uncertainties associated to the prediction of herbicide concentrations in streams.

Scientific and societal context

The occurrence of herbicides in surface and ground waters is a problem of great environmental concern. The effectiveness of mitigation and remediation measures relies on a better understanding of how the physical processes that control herbicide transport depend on agricultural practices or landscape characteristics. Although our intent is not to develop formal approaches to mitigate risk, our analysis is expected to improve the system knowledge which forms the basis to apply such techniques.

Keywords

Hydrological modelling, water quality modelling, herbicides, diffuse pollution, hypothesis testing, flexible frameworks, fieldwork
Direct link to Lay Summary Last update: 01.02.2016

Responsible applicant and co-applicants

Employees

Name Institute

Publications

Publication
Characterizing fast herbicide transport in a small agricultural catchment with conceptual models
Ammann Lorenz, Doppler Tobias, Stamm Christian, Reichert Peter, Fenicia Fabrizio (2020), Characterizing fast herbicide transport in a small agricultural catchment with conceptual models, in Journal of Hydrology, 586, 124812-124812.
A likelihood framework for deterministic hydrological models and the importance of non-stationary autocorrelation
Ammann Lorenz, Fenicia Fabrizio, Reichert Peter (2019), A likelihood framework for deterministic hydrological models and the importance of non-stationary autocorrelation, in Hydrology and Earth System Sciences, 23(4), 2147-2172.

Awards

Title Year
Outstanding Student Poster and PICO (OSPP) Award contest at the EGU General Assembly 2019 For: EGU2019-10662 Combining conceptual transport models with a new likelihood framework to assess high-frequency measurements of in-stream herbicide concentration by Lorenz Ammann et al. 2019

Abstract

Agriculture is one of the main sources of diffuse pollution and the occurrence of herbicides in surface and ground waters is one of the main undesired effects of agricultural practices. A detailed understanding of herbicide transport processes is crucial to the design of efficient mitigation measures. These processes have been studied extensively through in situ and laboratory experiments under relatively well-defined conditions. The transfer of this knowledge to the catchment scale, however, remains a major challenge.Catchment models are essential tools to investigate the system behaviour and plan efficient approaches for water protection. Models based on small scale physics are not readily applied at the scale of catchments and beyond, due to their large data requirement and computational costs. For this reason, models designed to directly represent emergent properties of the system are usually applied. Such models, however, are often criticized for their lack of theoretical rigour and limited predictive abilities. The lack of spatial data at the catchment scale is often advocated as a justification for the poor understanding of catchment scale processes. However, even when data is available, limited guidance exists on how to integrate it into the modelling process.In order to improve the representation of diffuse pollution processes at the catchment scale and their associated uncertainties, this project proposes to combine three approaches of model building: (1) Development of systematic approaches to synthesize experimental knowledge and translate it into information useful for modelling, such as qualitative perceptual models, and constraints on parameters, processes and states. (2) Use of a multi-model approach to find appropriate spatial representations of small catchments for simulating diffuse pollution from agricultural land, that account for the relevant spatial process chains, comply with available data, and are computationally feasible. (3) Consideration of input and intrinsic uncertainty and stochasticity by formulating model parameters as stochastic processes. This approach considers stochasticity without violating mass balances and leads to constructive suggestions for structural improvements of the model. In addition, we will investigate temporal and spatial model transferability.The modelling work will focus on two experimental catchments in the Swiss plateau (the Ror and the Eschibach catchments), which have been the location of unique controlled herbicide experiments at the catchment scale. Experimental data include the amount of applied herbicides, high frequency measurements of herbicide concentrations in the soil and at different locations in the stream providing spatial data on loss rates, measurements of groundwater levels and soil moisture, detailed soil maps, and laboratory analyses of herbicides sorption and decay properties.The expected outcomes of this project consist of (1) a better understanding of the contribution of experimental data in informing the modelling process of diffuse pollution from small agricultural catchments, (2) an improved understanding of the dominant controls on herbicide transport and of appropriate conceptualizations of catchment scale processes and (3) an assessment of the uncertainties associated to the prediction of herbicide concentrations in streams.
-