Assessment of Rainfall Influence Over Water Quality at the Effluent of an Urban Catchment by High Temporal Resolution Measurements

El presente estudio se trazó como objetivo establecer la influencia de la lluvia sobre la calidad del agua en el efluente de una cuenca urbana mediante el uso de medición en alta resolución temporal. Los parámetros de calidad y cantidad de agua en el efluente de la cuenca urbana estudiada (Gibraltar...

Descripción completa

Autor Principal: Sandoval Arenas, Santiago
Formato: masterThesis
Publicado: Pontificia Universidad Javeriana 2015
Materias:
The present study aimed to establish the rainfall influence over the water quality at the effluent of an urban catchment by the use of temporal high resolution measurements. The water quantity and quality parameters at the effluent of the studied urban catchment (Gibraltar: Bogotá, Colombia) were measured by means of infra-red, ion sensitive, UV-VIS spectrometry (water quality) and ultrasonic (water quantity: flow rate) sensors. Hence, the rainfall pulses were registered at three rain gauge stations (Fontibon, Bosa and Casablanca) in proximities of the studied urban catchment. A methodology for identifying periods in which water quality and quantity seemed to be ranging outside of their typical dry weather behavior was designed, by the use of signal processing and time series techniques (Median Filter, Step Detection and Autocorrelation functions-ACF-). A Principal Components Analysis (PCA) brought up a preliminary insight of which water quality and quantity parameters seem to explain mostly the differences between the variability of dry and wet weather data. The most explicative parameters measured by means of uni-parametric and ultrasonic sensors were: (i) the Dissolved Oxygen (DO), (ii) the Flow Rate, (iii) the Oxidation Reduction Potential (ORP), (iv) the Turbidity and (v) the Total Suspended Solids (TSS). The pH and the Temperature seemed not to be altered from dry to rainy weather. For the case of measurements undertaken by UV-VIS spectrometry, the absorbance values registered by the boundary wavelengths of the UV-VIS range (200 nm to 750 nm) seems to explain better the differences between variability of dry and wet weather days data, rather than for the wavelengths closer to the center of the UV-VIS range (for about 400 nm). Hence, considering the most explicative water quality and quantity parameters from PCA, the designed methodology was then applied to the time series registered during rainy days. A temporal delimitation of the rainfall events was obtained, which exposed an appropriate consistency. Regarding the identified rainfall events, association metrics (Pearson s Correlation Test, Average Mutual Information, Linear Models and Partial Least Squares Model) between different rainfall and water quality and quantity statistical characteristics per event (e.g. mean, standard deviation, entropy and duration) were undertaken, with the purpose of detecting recurrent relations between rainfall and water quality and quantity (for rainfall registered at different rain gauge stations). The results showed that: (i) Turbidity seems to be related with the duration of the rainfall pulses and with the mean rainfall height, (ii) DO seems to have an association with total volume of rainfall and with the rainfall pulse s entropy, (iii) flow rate was found to be related with mean rainfall height, standard deviation of rainfall heights and the mean rainfall intensity, (iv) TSS behavior was found to be linked with characteristics of rainfall such as the duration, the mean rainfall height, the standard deviation of the rainfall heights and the memory of the process (evaluated by the ACF of each specific rainfall pulse), (v) ORP seemed to be connected with the characteristics of rainfall events such as the duration, the Antecedent Dry Weather Period (ADWP), the total rainfall volume, the entropy and the standard deviation of the rainfall pulse(s). Rainfall registered at the Bosa rain gauge station seems to be more correlated with characteristics of the effluent in terms of quality and quantity, rather than Fontibon and Casablanca rain gauge stations. Preliminary insights showed that the principal sources of variability between the spatial influence of a given rain gauge station (in terms of the amount of recurrent) with different sub basins of the urban catchment are: (i) the distance to the rain gauge station which is mostly influencing a specific sub basin, (ii) the variability of the pipes slope at a given sub basin, (iii) the industrial land use regarding a sub basin, (iv) the mean pipes slope for a given sub basin and (v) the variability of pipes length at a sub basin.
Acceso en línea: http://hdl.handle.net/10554/12668
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares Similares