i think that the using of modflow is very useful for you , you will need the topographic map and the base layers elevation, soil characteristics, well system and different boundary condition, and some observation points for calibration process.
I am agree with Ramadan. Think that you should know soil characteristics and uses, drainage channels presence, proximity to channels, sea, with the object to avoid contaminants, salts intrusion .
An urban aquifer are ruled by the same hidrogeological basic parameters and has a general behaviour than a no urban one ( permeability, recharge rate, storage coefficient, etc. ). The main difference are the spatial distribution and intensity of the recharge, the quality of these water recharge, and the hidrological relationships between the aquifer and the urban structures that crosses their saturated zones, who provoke in some cases drainages or barrier effects. Codes like feflow or modflow (and many others) have numerical tools to incorporate all these factors in a flow or transport model. if you are interested to know more about these factors and the numerical reproduction of their effects I have the abstract of spanish congress that shows good examples in Barcelona city and other parts of Spain ( but the texts are in Spanish)
MODFLOW 2000 manual is located at http://water.usgs.gov/nrp/gwsoftware/modflow2000/ofr00-92.pdf
MODFLOW 2005 manual is located at http://pubs.usgs.gov/tm/2005/tm6A16/PDF.htm
Both have example problems you can try.
Also, if you use a GUI, such as Groundwater Vistas, Visual Modflow, GMS, etc. these software packages have tutorials in their respective manuals.
One other concern in modeling an urban area besides all the great suggestions above is the location of sewer lines. They can cause localized groundwater mounding due to leaking.