Improved understanding and prediction of the hydrologic response of highly urbanized catchments through development of the Illinois urban hydrologic model (IUHM)

Cantone, Joshua P.

Permalink

https://hdl.handle.net/2142/16073 Copy

Description

Title

Improved understanding and prediction of the hydrologic response of highly urbanized catchments through development of the Illinois urban hydrologic model (IUHM)

Author(s)

Cantone, Joshua P.

Issue Date

2010-05-19T18:34:00Z

Director of Research (if dissertation) or Advisor (if thesis)

Schmidt, Arthur R.

Doctoral Committee Chair(s)

Garcia, Marcelo H.

Committee Member(s)

• Schmidt, Arthur R.
• Kumar, Praveen
• Sivapalan, Murugesu
• Tung, Yeou-Koung

Department of Study

Civil & Environmental Eng

Discipline

Civil Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Urban hydrology
• uncertainty
• dispersion mechanisms
• Geomorphologic Instantaneous Unit Hydrograph (GIUH)
• modeling

Abstract

What happens to the rain in highly urbanized catchments? That is the question that urban hydrologists must ask themselves when trying to integrate the hydrologic and hydraulic processes that affect the hydrologic response of urban catchments. The Illinois Urban Hydrologic Model (IUHM) has been developed to help answer this question and improve understanding and prediction of hydrologic response in highly urbanized catchments. Urban catchments are significantly different than natural watersheds but there are similarities that allow features of the pioneering GIUH concept developed for natural watersheds to be adapted to the urban setting. This probabilistically based approach is a marked departure from the traditional deterministic models used to design and simulate urban sewer systems, and does not have the burdensome input data requirements that detailed deterministic models possess. Application of IUHM to the CDS-51 catchment located in the Village of Dolton, IL highlights the models ability to predict the hydrologic response of the catchment as well as the widely accepted SWMM model and in accordance with observed data recorded by the USGS. The model is further used to improve the understanding of urban catchment hydrology. It is shown that inlet storage and pressurized flow can have a significant impact on the hydrologic response in urban catchments. In addition, the unique structure and organization of urban sewer networks make it possible to characterize Horton’s Laws in urban catchments. Overall, the results provide invaluable insight into how the different hydrologic/hydraulic processes encountered in urban catchments effect the hydrologic response of the catchment. The link between river network structure and hydrologic response for natural watersheds has been the subject of ongoing research for the past 30 years. In this research we investigate the link between sewer network structure and hydrologic response in urban catchments. It has been shown in natural watersheds that there are dispersion mechanisms that contribute to the impulse response function of the catchment: hydrodynamic dispersion, geomorphologic dispersion and hydrodynamic dispersion. We introduce a fourth dispersion mechanism, intra-state dispersion, that accounts for the variance in conduit (e.g. slope, length, diameter etc.) and overland region input parameters (e.g. slope, area, imperviousness etc.) within an order. This dispersion mechanism is found to be the second largest contributor to the total dispersion in urban catchments, contributing less than hydrodynamic dispersion, but more than kinematic and geomorphologic dispersion. Furthermore, an uncertainty analysis is performed to help better understand the uncertainty in the predicted hydrologic response that is introduced by spatial variation in conduit and overland input parameters. It is identified that conduit slope and length are the greatest sources of uncertainty in the predicted direct runoff hydrograph for the CDS-51 catchment in the Village of Dolton, IL, and the CDS-36 catchment in the City of Chicago, IL. The IUHM requires, as input, the mean and variance of parameters including conduit slope, overland slope, subcatchment area and imperviousness. Ideally the first two statistical moments of each of these input parameters would be calculated using the deterministic data for all conduits and subcatchments in the system. In reality, such detailed information is not always available, is uncertain, or time cannot be afforded to delineate all of the sub-areas within the catchment. IUHM was designed so that it could be used in such situations under the hypothesis that the mean and variance of the input parameters could be determined using only a sub-set of the full deterministic dataset. This hypothesis is tested and it is shown that a random sample capturing as little as 30% of the subcatchments and conduits in the CDS-51 catchment (located in the Village of Dolton, IL) can be used to generate the mean and variance in the ith-order conduit slope, overland slope, subcatchment area and imperviousness without significantly reducing the accuracy or increasing the uncertainty of the predicted hydrologic response. IUHM provides an alternative to traditional deterministic models that maintains the non-linearity in the key physical processes at the urban scale and is capable of accounting for the uncertainty caused by spatial variation in the input parameters throughout the catchment. This model is still in its infancy and as such has the potential to be improved through ongoing research. The model has been setup to allow other hydrologic and hydraulic processes (e.g. stormwater best management practices, dual drainage) to be easily incorporated into the model.

Graduation Semester

2010-5

Permalink

http://hdl.handle.net/2142/16073

Copyright and License Information

Copyright 2010 Joshua P. Cantone

Owning Collections