Answers are grounded in this paper and cite their sources.
Rainfall is the most important hydrological attribute on which the socio-economic development and agricultural productivity of India depend strongly. Rainfall is highly variable spatially and temporally and depends on various meteorological and terrain parameters. Statistical models are commonly used to predict rainfall based on time series data. Artificial neural networks (ANNs) are found to be far more superior and accurate in modelling the relationship between inputs and outputs considering multiple and diverse variables without explicit physical consideration. In the present studies, (1) the spatial and temporal variability of rainfall and the parameters is analysed and (2) the merits of ANNs are taken advantage of to predict spatial distribution of rainfall over the peninsular India utilising the meteorological and physical parameters. The data on these parameters obtained from different sources with different formats and resolutions was brought to a common format, resolution and size for use with ANNs. It was found that ANNs not only permit diverse multi-thematic input data but also predict spatial distribution of rainfall far more accurately.
No takes yet. Share an insight, caveat, or question.
Richard et al. (2013) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: