The Learning of Multivariate Adaptive Regression Splines (MARS) Model in Rainfall-Runoff Processes at Pahang River Catchment
-
Published:2018-10-01
Issue:2
Volume:18
Page:161-167
-
ISSN:2393-1493
-
Container-title:Annals of Valahia University of Targoviste, Geographical Series
-
language:en
-
Short-container-title:
Author:
Halid D.A.1, Atan I.1, Jaafar J.1, Ashaari Y.1, Mohamed S.N.2, Samsudin M.B.3, Baki A.3
Affiliation:
1. Faculty of Civil Engineering , Universiti Teknologi MARA , Shah Alam , Selangor, Malaysia 2. Faculty of Civil Engnineering , Universiti Teknologi MARA , Cawangan Johor Kampus Pasir Gudang, Johor, Malaysia 3. Envirab Services, P.O.Box 7866, GPO Shah Alam 40730 , Selangor, Malaysia
Abstract
Abstract
Recently, a novel data mining technique, Multivariate Adaptive Regression Splines (MARS) has begun attracted attention from several hydrological researchers because their application is relatively new in modelling hydrological processes. The power of this approach has been proven in variety learning problems such as financial analysis, species distributions modelling, and doweled pavement performance modelling. Therefore, the objective of this paper is to investigate the performance of MARS model in capture the rainfall-runoff processes at river catchment of Malaysia. Pahang River has been selected as area of study. 30-years data set of daily rainfall and runoff at upstream tributaries of Pahang River were used to developed and validate the capability of MARS model in flood prediction. The effect of different length of record data to performance of MARS model was also examined by arranged the data into 5-years data set, 10 years data set, 20 years data set, and 30 years data set. All these data sets used 1-year data of 2003 for validation process while the others were applied for calibration. Simulation results showed that MARS model was able to learn the rainfall-runoff processes in Pahang River catchment and the model performance improved due to the longer period of data.
Publisher
Walter de Gruyter GmbH
Reference16 articles.
1. Abraham, A., Steinberg, D., (2001a), MARS: still an alien planet in soft computing? Int. conf. Comput. Sci.2, 235-244. 2. Abraham, A., Steinberg D., (2001b), Is neural network a reliable forecaster on earth? A MARS query! IWANN 2, 679-686. 3. Ames, D.P. (1998), Seasonal to Interannual Streamflow Forecasts Using Nonlinear Time Series Methods and Climate Information, M.S thesis, Utah State University, Logan, Utah. 4. Briand, L.C., Freimut, B., and Vollei, F., (2004), Using Multiple Adaptive Regression Splines to Support Decision Making in Code Inspections, The Journal of Systems and Software, 73: 205-217 5. DID (Department of Irrigation and Drainage Malaysia), (2003), Volume 2: Updating of Condition of Flooding in Malaysia, Annual Report 2003, Malaysia.
|
|