Predicting ¯oods from urban development scenarios: case study of the Dil uvio Basin, Porto Alegre, Brazil N estor A. Campana a, * , Carlos E.M. Tucci b a Civil Engineering Department, University of Bras õlia, University Campus, North Wing, 70910-900 Bras õlia, DF, Brazil b Institute of Hydraulic Research, Federal University of Rio Grande do Sul, Av. Bento Gonc ßalves, 9500, Postal code 15029, 9501-970, Porto Alegre, RS, Brazil Received 30 July 1999; received in revised form 28 November 2000; accepted 1 February 2001 Abstract It is well known that urbanization increases the hydrograph peak and overland ¯ow volume from drainage basins, and that the concentration time of ¯ood ¯ows is diminished. To predict the eects of future urban development on the ¯ood regime, the design hydrograph must be estimated. Nevertheless, the information from which to estimate a design hydrograph is limited where planning procedures such as the `Urban Master Plan' of some Brazilian cities) specify only the type of urban land use residential, industrial, etc.) and the recommended population density in areas where development is planned. This paper presents the relationship between parameters of a hydrologic model and urban development characteristics as set out in the Urban Master Plan. A hydrologic model IPH IV [C. E. M. Tucci, B. Braga, A. L. L. Silveira, RBE Caderno de Recursos H õdricos 7 1) 1989)] was used together with a GIS to predict the hydrograph corresponding to alternative urbanization scenarios. Model parameters were ®tted and veri®ed using recorded data from the Dil uvio Creek in Porto Alegre, Brazil. The model was applied to predict the design hydrograph for dierent risks in accordance with the city's Urban Master Plan. Flooded areas corresponding to a storm rainfall with return period 25 years were calculated by simulation, the ¯ooding being mainly due to ¯ow obstruction by bridges. Ó 2001 Elsevier Science Ltd. All rights reserved. Keywords: Design hydrograph; Urban development; GIS 1. Introduction Almost 80% of Brazil's total population live in cities. There are more than 12 cities with population greater than 1 million and many others of medium size where the rate of population increase is high. The high popu- lation density in cities has had a major eect on urban drainage since there is no regulation for control of peak ¯ow at source. Floods occur ever more frequently and increased ¯ood ¯ows cause even greater public damage, while the solution adopted has generally been to increase channel conveyance, at high cost and low eciency. The cost of controlling ¯ood drainage after areas have been urbanized is so high that most municipal districts do not have sucient funds to build the nec- essary protection works. More ecient, preventive control is now being accomplished by predicting the impacts from the potential urban developments, and planning control measures at an early stage. These forecasts are incorporated into the Urban Master Plan, thus minimizing future damage and controlling costs. It is well known that urbanization within a drainage basin tends to increase peak discharge, to decrease the time taken for the ¯ood discharge to reach its peak, and to increase the volume of runo. Predictions of the hydrograph to be expected following major urban de- velopment are usually obtained by rainfall-runo mod- eling, and several such models have been used SCS, 1975; Crawford & Linsley, 1966). To obtain a reliable prediction, the following are needed: a) the design rainfall for the speci®ed intensity, duration and return period, b) basin characteristics such as drainage area, stream length, soil type, natural vegetation cover; c) urbanization characteristics, such as drainage network geometry and structure, and locality and extent of im- permeable areas. Urban drainage planning is a process that should begin whilst a basin is still in its natural state and before urbanization has been initiated. The work reported in Urban Water 3 2001) 113±124 www.elsevier.com/locate/urbwat * Corresponding author. E-mail addresses: mnestor@unb.br N.A. Campana), tucci@ if.ufrgs.br C.E.M. Tucci). 1462-0758/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved. PII:S1462-075801)00004-8