Inspite of all the expertise deployed in data-collection, modeling and inference in which a large number of talented, skilled and well trained brains are employed for doing the job with high-tech communication and computational instruments and networking, it is amazing to look at the natural capabilities of several species of birds and animals in sensing the future weather and geo-events far more precisely than by the well-known experts using high-tech instruments. Meteorology departments or weather bureaus in most parts of the world are now well equipped with sophisticated satellite based data collection devices, geo-information systems and remote sensing devices to continuously record the signals and signs of the variations deep under the sea, beneath the earth and far away in the space. However, the multiple dimensions of parameters, random nature and complexity of the data make the analysis difficult and so precise and reliable inference models and accurate predictions continue to elude from their sight. Perhaps this is the reason people often joke that – "If the weather bureau predicts heavy rains, one should better not bother to take any umbrella or raincoat with him, because the atmosphere is known to not only 'disobey' but also 'contradict' their estimates". Electronic media keeps spreading the 'predictions' of the local and global weather departments throughout the day so there is no dearth of information. But the question is how reliable that information is? No doubt, the accuracy of predictive methods has improved significantly over the past two decades, still the lack of precision and time-lapses witnessed in the tragic occurrences of the recent calamities of earthquakes in Gujarat, Indonesia, Tsunami at south-eastern Indian coast, Katrina in the USA, necessitates the need for further advancement of science and technology to face Nature's challenges with timely preparations. The present trends indicate that the predictions using modern technological gadgets are correct on an average scale for more-or-less periodic events. For example, the overall nature of monsoon in a given year generally is in good agreement with what has been inferred a month or fortnight before its onset. Prominent among the geological events that are prone to bring a drastic environmental shift for South Asia are – El Nino (associated with the temperature of the Pacific ocean in regions towards coasts of Peru, Latin America) and the pressure belt from Darwin (Australia) to Argentina. Correct inference of monsoon plays a key role in national and global economy. It is more important for countries like India, where the state of agricultural production controls the gross economic growth and this crucial sector largely depends upon monsoons for irrigation. In terms of the best-validated models so far, the inference of monsoon status is made based on estimation of sixteen parameters. Six of these deal with temperature variation pattern, two with snowfall, two with the speeds and directions of winds and five with the patterns of atmospheric pressure in specific regions. Estimators of these based on the current season and previous few years' data are analyzed to assess the 'state of monsoon' in general. Thousands of scientists and data processors are required to work hard for several months, using high-performance computers with specialized statistical and mathematical software for this analysis. In broad terms, if the model fitting and simulation shows indices based on all of these parameters as satisfactory then the monsoon is predicted as "average". Similar criteria satisfied by ten to twelve of these indicate above average rainfall during the monsoon. More that twelve of these being favorable implies very good monsoon. The statistics of the Meteorological Department of India support the validity of these criteria. In the year 1988, fourteen of these indices were found favorable. The monsoon that year proved the simulations right, as the rainfall was higher than the record of past fifty years.

- courtesy Akhand Jyoti