University of Twente Student Theses

Login

Stochastenmethode in regionaal waterbeheer: Het beoordelen of met de stochastenmethode een goede indicatie van normafvoeren kan worden gegeven door gebruik te maken van een hydrologisch Simgro-model

Meurs, Sebastiaan (2013) Stochastenmethode in regionaal waterbeheer: Het beoordelen of met de stochastenmethode een goede indicatie van normafvoeren kan worden gegeven door gebruik te maken van een hydrologisch Simgro-model.

[img] PDF
14MB
Abstract:The Dutch waterboards face with the objective to verify the safety of regional water systems by controlling the safety standards. Heavy rainfall events during the 90s showed the vulnerability of the regional water systems and gave rise to the objective. The safety standards vary from return periods of once per year till once per hundred years. The variation of safety standards is the result of different social and economical values that the landsurface can perform. When safety standards are exceeded, the Dutch waterboards are assigned to restore the regional watersystem before 2021. The Nationaal Bestuursakkoord Water (2003) prescribes that safety standards can be quantified by applying the time series methodology or the stochastic methodology. Both methods are based on statistics. In case of the time series methodology, the safety standards are quantified after modeling a long historical rainfall period. The main principle of the stochastic methodology, a relative new methodology in the field of water management, is that extreme water levels/discharges can be explained by different hydrological and meteorological circumstances. The circumstances are called stochasts. Important stochasts are identified first. Each stochast can perform different conditions (different rainfall volumes for example). The probability of the conditions is based on statistics of extreme rainfall. Stochastic events are obtained by combining conditions of the different stochasts. Because of the assumption that the conditions of the different stochasts are independent, the probability of an event equals the product of the probabilities of the combined conditions. All of the events are simulated. The simulated discharges are used to quantify the safety standards. The Dutch waterboard Peel en Maasvallei (WPM) wants to quantify the safety standards with the stochastic methodology. Added to the quantification of the safety standards with the stochastic methodology, WPM wants to determine the safety standards by applying the currently used hydrological Simgro- model. The research objective of this study is comparable and is stated by the examination whether the stochastic methodology may be appropriatie in the determination of safety standards by using the hydrological Simgro- model. The current Simgro- model shows a bad model performance. Peak discharges are underestimated (by factor two) and the observed average discharge downstream is underestimated with about 30%. To quantify the safety standards, the simulation of peak discharges is the primarly concern. As result of that, a plan to improve the model performance is initiated. The plan contains several model adjustments. The simulation of the model adjustments show that both the underestimation of the average discharge and peak discharges are reduced. However, the underestimation of peak discharges is still 20-30%. Due to the high importance of modeling peak discharges, the underestimation of peak discharges is still relatively large. Despite of that, it is decided to accept it for this study. The stochastic methodology is conducted by considering rainfall related stochasts (rainfall volume and intensity) and the initial saturation condition of the soil. Due to the varying features of rainfall and the variation of saturation conditions per season, two seasons are considered; the growing season (March- October) and the winter (November-February). The rainfall duration is also considered, but it is decided to simplify this stochast. The simplification emerges by considering one rainfall duration per season. Calculations, concerning the determination of the rainfall pattern, make clear that one-day events are prominent for events during the growing season and four-day events for the winter. The total number of stochastic events, which are simulated with the improved Simgro- model, is 182. When the quantified safety standards are compared to the observations it is made clear that the simulations show lower discharges. The difference is 12-20%. The results of the simulation of the stochastic events also show that high discharges are mainly simulated during the growing season. The importance of the saturation condition of the soil is in according the simulation relatively small. The limited added value of winter events and the saturation of the soil is in contrast with the expectations. Observed rainfall events show that high discharges mainly occur during the winter and are caused by a saturated soil. The simulated events show that the quantified safety standards based on the observations are higher. The difference may be forced by the way the infiltration capacity is incorporated to consider the surface runoff. The applied infiltration capacity is stationary and does not consider the saturation of the soil. Applying a non- stationary infiltration capacity, which is dependent on the soil saturation, may improve the simulation of peak discharges. Secondly, reducing the drainage resistance may, caused by the faster runoff, provide better simulations of peak dicharges. The simulation of lower discharges may also be induced by the applied discretisation of the initial soil saturation during winter events. When incorporating a wetter soil condition during the winter, it is expected that the quantified safety standards based on simulations are more comparable to the quantified safety standards based on observations. Added tho these recommendations, it is also advised to do more investigation to the dependancy between stochasts. The research objective may be answered cautiously positive. By using the Simgro- model, it is not possible to quantify safety standards currently, but executing the recommendations may possibly remove doubts about the stochastic methodology in combination with the Simgro- model.
Item Type:Essay (Master)
Faculty:ET: Engineering Technology
Subject:56 civil engineering
Programme:Civil Engineering and Management MSc (60026)
Link to this item:http://purl.utwente.nl/essays/64467
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page