Loading...

Forecasting Urban Groundwater Level Applying Geographical Information System (GIS) and Artificial Neural Network (ANN)

Jazaei, Farhad | 2009

906 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 39323 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Abrishamchi, Ahmad
  7. Abstract:
  8. Groundwater beneath the cities is becoming an important and valuable resource. Conjunctive use of surface and groundwater is likely to become increasingly more common as urban population grows by time. Therefore, one important requirement for urban water management planning is forecasting the groundwater level fluctuations. Unfortunately less experience and information is available to evaluate the fluctuations of groundwater level in urban environment compare to the natural systems, also different processes (sources) are involved in an urban water cycle, which all together make it more complicated to study. Similar to many other megacities, there is a serious lack of hydrogeological and long-period time-series data in a megacity like Tehran, Iran, which is mostly due to the limitations of allocated budget and complexity of data gathering processes. Urban groundwater fluctuation process is so complicated and any research in this field needs to gathering a lot of hydrological and hydrogeological data in a vast extent of a city or megacity. So the main objective of this research was forecasting the urban groundwater level by using common and available hydrological data such as groundwater level, precipitation, temperature and in-city stream flow time-series instead of more scarce data such as hydrogeological data. Urban groundwater level fluctuation should have studied in a distributed manner. So in this research, Geographical Information System (GIS) tool has been used to handling the temporal and spatial data and making a sufficient data-bank. Finally these gathered data has been processed by the intelligent Artificial Neural Network (ANN) models to predicting the urban groundwater level. In this research, we investigate that: (1) making an appropriate data-bank applying Geographical Information System (GIS) can increase the accuracy of ANN model’s forecasting and sufficiency of distributed studying of urban groundwater level fluctuation, (2) ANN models have great capability to predict the urban groundwater level using different sets of available input data. The accuracy of the prediction can be increased by using proper input and lag time sensitive analysis for each pizometric wells and finally selecting the best input data set and input lag times, (3) Three learning algorithms (LM,GDX and BR) nearly have a same accuracy in which we can remark that LM is the best, and (4) ANN models with assistance of proper GIS data-bank (time series with 160 months) are capable in prediction of urban groundwater level in 1 and 2 months (steps) ahead an acceptable confidence level
  9. Keywords:
  10. Forecasting ; Artificial Neural Network ; Geographic Information System (GIS) ; Water Level ; Groundwater

 Digital Object List

 Bookmark

No TOC