Integrating hydrometeorological information for rainfall‐runoff modelling by artificial neural networks

Abstract

Abstract:The major purpose of this study is to effectively construct artificial neural networks-based multistep ahead flood forecasting by using hydrometeorological and numerical weather prediction (NWP) information. To achieve this goal, we first compare three mean areal precipitation forecasts: radar/NWP multisource-derived forecasts (P r ), NWP precipitation forecasts (P n ), and improved precipitation forecasts (P m ) by merging P r and P n . The analysis shows that the accuracy of P m is higher than that of P r and P n . The analysis also indicates that the NWP precipitation forecasts do provide relative effectiveness to the merging procedure, particularly for forecast lead time of 4-6 h. In sum, the merged products performed well and captured the main tendency of rainfall pattern. Subsequently, a recurrent neural network (RNN)-based multistep ahead flood forecasting techniques is produced by feeding in the merged precipitation. The evaluation of 1-6-h flood forecasting schemes strongly shows that the proposed hydrological model provides accurate and stable flood forecasts in comparison with a conventional case, and significantly improves the peak flow forecasts and the time-lag problem. An important finding is the hydrologic model responses which do not seem to be sensitive to precipitation predictions in lead times of 1-3 h, whereas the runoff forecasts are highly dependent on predicted precipitation information for longer lead times (4-6 h). Overall, the results demonstrate that accurate and consistent multistep ahead flood forecasting can be obtained by integrating predicted precipitation information into ANNs modelling.

Flood and groundwater management for the mountain plateau of Omalos based on geoinformatics techniques

“…Thus, in flash floods studies, precipitation records and flow data have been widely used (Chiang and Chang 2009); however, the use of systematic data on flash floods presents several challenges in mountainous catchments as representative instrumental records are not normally available in these environments (Ballesteros et al 2010). Thus, in ungauged mountainous areas where expensive and time consuming hydrological-hydraulic simulations are not possible the use of an effective Geographic Information Systems (GIS) management tool is essential to delineate the flood prone areas (Manfreda et al 2008).…”

Dynamic factor analysis and artificial neural network for estimating pan evaporation at multiple stations in northern Taiwan

“…In recent decades, a number of neural networks, such as the back-propagation neural network (Rumelhart et al 1986), the recurrent neural network (Williams 1989) and the fuzzy neural network (Nie and Linkens 1994), were developed to solve a wide variety of problems (Schalkoff 1997, Chang et al 2008, Chen and Chang 2009, Chiang and Chang 2009, Ozkan et al 2011. Researchers tried to estimate evaporation by fitting the relationship between meteorological factors (Burman 1977, Gavin and Agnew 2004, Kisi, 2006.…”

Including spatial distribution in a data-driven rainfall-runoff model to improve reservoir inflow forecasting in Taiwan

“…The result is a dominance of spatially lumped data‐driven studies, and this is especially true in the case of rainfall‐runoff models (e.g. Nayak et al ., ; Chiang and Chang, ; Wu and Chau, ; De Vos, ). Despite its obvious potential as a means by which spatial variation can be captured and incorporated, few data‐driven modelling studies have attempted to use raster‐based radar rainfall (e.g.…”