Experiments in Short-Term Precipitation Forecasting Using Artificial Neural Networks

Kuligowski, Robert J.; Barros, A. P.

doi:10.1175/1520-0493(1998)126<0470:eistpf>2.0.co;2

Cited by 115 publications

(45 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The value of the threat score varies from 0 (completely unrelated) to 1 (for a perfect forecast) (Kuligowski and Barros 1998). Because of large differences in the precipitation amounts at the stations located on the east and west face of the Sierra Nevada, the threshold values were taken as relative to the mean station precipitation.…”

Section: Model Performancementioning

confidence: 99%

A Hybrid Orographic plus Statistical Model for Downscaling Daily Precipitation in Northern California

Pandey¹,

Cayan²,

Dettinger³

et al. 2000

J. Hydrometeor

View full text Add to dashboard Cite

A hybrid (physical-statistical) scheme is developed to resolve the finescale distribution of daily precipitation over complex terrain. The scheme generates precipitation by combining information from the upper-air conditions and from sparsely distributed station measurements; thus, it proceeds in two steps. First, an initial estimate of the precipitation is made using a simplified orographic precipitation model. It is a steady-state, multilayer, and two-dimensional model following the concepts of Rhea. The model is driven by the 2.5Њ ϫ 2.5Њ gridded National Oceanic and Atmospheric Administration-National Centers for Environmental Prediction upper-air profiles, and its parameters are tuned using the observed precipitation structure of the region. Precipitation is generated assuming a forced lifting of the air parcels as they cross the mountain barrier following a straight trajectory. Second, the precipitation is adjusted using errors between derived precipitation and observations from nearby sites. The study area covers the northern half of California, including coastal mountains, central valley, and the Sierra Nevada. The model is run for a 5-km rendition of terrain for days of January-March over the period of 1988-95. A jackknife analysis demonstrates the validity of the approach. The spatial and temporal distributions of the simulated precipitation field agree well with the observed precipitation. Further, a mapping of model performance indices (correlation coefficients, model bias, root-mean-square error, and threat scores) from an array of stations from the region indicates that the model performs satisfactorily in resolving daily precipitation at 5-km resolution.

show abstract

Section: Model Performancementioning

confidence: 99%

A Hybrid Orographic plus Statistical Model for Downscaling Daily Precipitation in Northern California

Pandey¹,

Cayan²,

Dettinger³

et al. 2000

J. Hydrometeor

View full text Add to dashboard Cite

show abstract

“…For example the floods between November 2010 and January 2011 that left three-quarters of Queensland, Australia a disaster zone (Hurst, 2011) were not predicted well in advance (Abbot and Marohasy, 2014;Inquiry, 2011;Seqwater, 2011). Despite improvements in the performance of numerical weather models, they do not provide quantitative precipitation forecasts at enough spatial and temporal scales (Kuligowski and Barros, 1998). Consequently there is a gap in rainfall prediction capability by the GCM especially beyond 1 week or shorter than a season (Hudson et al, 2011).…”

Section: Introductionmentioning

confidence: 99%

Application of the extreme learning machine algorithm for the prediction of monthly Effective Drought Index in eastern Australia

Deo

Şahin

2015

Atmospheric Research

255

102

View full text Add to dashboard Cite

The prediction of future drought is an effective mitigation tool for assessing adverse consequences of drought events on vital water resources, agriculture, ecosystems and hydrology. Data-driven model predictions using machine learning algorithms are promising tenets for these purposes as they require less developmental time, minimal inputs and are relatively less complex than the dynamic or physical model. This paper authenticates a computationally simple, fast and efficient non-linear algorithm known as extreme learning machine (ELM) for the prediction of Effective Drought Index (EDI) in eastern Australia using input data trained from 1957-2008 and the monthly EDI predicted over the period 2009-2011. The predictive variables for the ELM model were the rainfall and mean, minimum and maximum air temperatures, supplemented by the large-scale climate mode indices of interest as regression covariates, namely the Southern Oscillation Index, Pacific Decadal Oscillation, Southern Annular Mode and the Indian Ocean Dipole moment. To demonstrate the effectiveness of the proposed datadriven model a performance comparison in terms of the prediction capabilities and learning speeds was conducted between the proposed ELM algorithm and the conventional artificial neural network (ANN) algorithm trained with Levenberg-Marquardt back propagation. The prediction metrics certified an excellent performance of the ELM over the ANN model for the overall test sites, thus yielding Mean Absolute Errors, Root-Mean Square Errors, Coefficients of Determination and Willmott's Indices of Agreement of 0.277, 0.008, 0.892 and 0.93 (for ELM) and 0.602, 0.172, 0.578 and 0.92 (for ANN) models. Moreover, the ELM model was executed with learning speed 32 times faster and training speed 6.1 times faster than the ANN model. An improvement in the prediction capability of the drought duration and severity by the ELM model was achieved. Based on these results we aver that out of the two machine learning algorithms tested, the ELM was the more expeditious tool for prediction of drought and its related properties.

show abstract

“…Rainfall forecasting can apply to many time horizons such as short term [3], medium term, and long term periods [4] [5]. Some authors design systems which can forecast yearly data, some try to forecast monthly data [5] whereas some try to forecast daily data [6].…”

Section: Related Workmentioning

confidence: 99%

Rainfall forecasting models using focused time-delay neural networks

Htike

Khalifa

2010

International Conference on Computer and Communication Engineering (ICCCE'10)

View full text Add to dashboard Cite

Abstract-Rainfall forecasting is vital for making important decisions and performing strategic planning in agriculturedependent countries. Despite its importance, statistical rainfall forecasting, especially for long-term, has been proven to be a great challenge due to the dynamic nature of climate phenomena and random fluctuations involved in the process. Artificial Neural Networks (ANNs) have recently become very popular and they are one of the most widely used forecasting models that have enjoyed fruitful applications for forecasting purposes in many domains of engineering and computer science. The main contribution of this research is in the design, implementation and comparison of rainfall forecasting models using Focused TimeDelay Neural Networks (FTDNN). The optimal parameters of the neural network architectures were obtained from experiments while networks were trained to perform one-step-ahead predictions. The daily rainfall dataset, obtained from Malaysia Meteorological Department (MMD), was converted to monthly, biannually, quarterly and monthly datasets. Training and testing were performed on each of the datasets and corresponding accuracies of the forecasts were measured using Mean Absolute Percent Error. For testing data, results indicate that yearly rainfall dataset gives the most accurate forecasts (94.25%). As future work, more parameters such as temperature, humidity and sunshine data can be incorporated into the neural network for superior forecasting performance.

show abstract

Experiments in Short-Term Precipitation Forecasting Using Artificial Neural Networks

Cited by 115 publications

References 0 publications

A Hybrid Orographic plus Statistical Model for Downscaling Daily Precipitation in Northern California

A Hybrid Orographic plus Statistical Model for Downscaling Daily Precipitation in Northern California

Application of the extreme learning machine algorithm for the prediction of monthly Effective Drought Index in eastern Australia

Rainfall forecasting models using focused time-delay neural networks

Contact Info

Product

Resources

About