Automated runoff coefficient computation in urban drainage systems using Google satellite images and fuzzy classification.

---

Neiler Medina, Arlex Sanchez, Zoran Vojinovic

Thursday 2 july 2015

16:00 - 16:15h at Central America (level 0)

Themes: (T) Water resources and hydro informatics (WRHI), (ST) Catchment hydrology

Parallel session: 13H. Water resources - Catchment

---

In the process of designing storm or combined water network systems, there are numerous and repetitive tasks that require not only substantial time for their execution but also they are very much dependent on judgment and experience of the design engineer. Such examples are catchment delineation, identification of runoff coefficients and definition of dry and wet weather flow parameters. Dealing with such tasks in real-life projects is time consuming. This paper provides a new method and tool to contribute in the automation of such tasks by evaluating satellite imagery obtained from Google Maps in an automated fashion and using fuzzy logic classification was possible to calculate or estimate the values of runoff coefficients for different subcatchments within the coverage of the satellite images. A graphical user interface (Runoff Extractor ver B1.0) was built using Delphi code environment; This GUI allow the final user to download the high resolution images from the educational license of the Google maps API from the desired area and then calculate the Runoff characteristics that are needed in order to build the input file for EPA SWMM model. The approach was tested on three case studies: the city of Birmingham (UK), Quito (Ecuador) and Medellín (Colombia), and the results obtained so far are promising when compared with the outcome of the calibrated models that already exist in the study areas.