Long Range Forecast for 2007 South-West Monsoon Season Rainfall

The long range forecast for the 2007 South-west monsoon (June to September) by the India Meteorological Department (IMD) says that the rainfall for the country as a whole is likely to be 95% of the long period average with a model error of ± 5%. This is based upon the newly-adopted statistical forecast system.

IMD will update the above forecast in June 2007 as a part of the second stage forecasts. Separate forecasts for the July rainfall over the country as a whole and seasonal (June-September) rainfall over the four geographical regions of India also will be issued.

India Meteorological Department has been adopting a two-stage forecast strategy for the south-west monsoon rainfall for the past four years. The first long range forecast for the south-west monsoon season (June-September) rainfall is issued in April using the 8 parameter models and the forecast update is issued in June based on the 10 parameter models.

IMD has been making consistent efforts to improve the long range forecasting system. IMD has now developed new statistical models for forecasting south-west monsoon rainfall (June – September) for the country as a whole, which are being introduced this year.

a) A new 5- parameter statistical forecasting system requiring data up to March, will replace the existing 8 parameter power regression model for the first forecast in April.

b) A new 6- parameter statistical forecasting system requiring data up to May will replace the existing 10 parameter power regression model for the forecast update in June.

There are three major changes in the new statistical forecast models from the 8/10 Parameter models. They are: a) a new smaller predictor data set b) use of a new non-linear statistical technique along with conventional multiple regression technique c) application of the concept of ensemble averaging.

A common weakness of all statistical models is that while the correlations are assumed to remain constant in future, they may, and in fact do, change with time and slowly lose their significance. IMD has embarked an exercise to examine the stability of the predictors and update the list of predictors. This exercise has yielded a new set of 8 predictors. For the April forecast, models are developed using a set of 5 predictors. For the updated forecast in June, models are developed using a set of 6 predictors, which includes 3 predictors used for the April forecast. The 8 predictors considered for the new ensemble forecast system are given below:

The most important aspect of the new forecast system is the introduction of the concept of ensemble forecasts. In this method, instead of relying on a single model, we have considered all the models with all possible combination of predictors. With 5 predictors, 31 different models are possible. Out of all possible models, the best few models were selected based on the skill in predicting monsoon rainfall during a common period. For the April forecast, 6 best models were identified. Ensemble mean is computed as the weighted average of the best six models thus identified. The weights are proportional to the skill of the models.

For developing the models, two different statistical techniques namely, Multiple Regression (MR) and Projection Pursuit Regression (PPR) were considered. While the MR technique is a conventional linear model technique and more commonly used, the PPR technique is a non-linear technique, used for the first time in forecasting Indian monsoon rainfall. The PPR method is known for its superiority in capturing the non-linear relationships between the predictors and rainfall. Verification of the results with the past data showed that the ensemble method performed better than the individual models.

The new statistical forecast system has also shown better performance compared to the 8 and 10 parameter models during the recent years, including the drought years of 2002 and 2004. The model errors of the April and June forecast systems however remain as ±5% and ±4% respectively.

The methods used in the new forecast system and results have been documented, peer reviewed and published in 2006 in a reputed international research journal, Climate Dynamics.

As a part of ongoing efforts to improve the long range forecast capabilities, experimental forecasts for the 2007 south-west monsoon rainfallbased on the IMD’s dynamical forecast system were also generated. For this purpose, observed sea surface temperature data of March have been used.

In addition, IMD has also taken into account the experimental forecasts prepared by national institutes like Indian Institute of Tropical Meteorology, Pune, Indian Institute of Science, Bangalore, Space Applications Centre, Ahmedabad, National Institute of Oceanography and Centre for Mathematical Modelling and Computer Simulation (CMMACS) Bangalore and operational/experimental forecasts prepared by international institutes like the National Centers for Environmental Prediction (NCEP), USA, International Research Institute for Climate and Society (IRI), USA, Meteorological Office, UK, the European Center for Medium Range Weather Forecasts(ECMWF), UK and the Experimental Climate Prediction Center (ECPC), USA.

During the end of August 2006, moderate El Nino conditions developed over the equatorial Pacific Ocean, but the event was very short lived. The warm sea surface temperature (SST) anomalies over the east equatorial Pacific have disappeared during February 2007. By the end of February, SSTs were near average in the vicinity of the date line, and below average over the eastern equatorial Pacific. The equatorial upper-ocean heat content (average temperature departures in the upper 300 m of the ocean) also decreased rapidly. These trends in surface and subsurface ocean temperatures indicate that the warm (El Niño) episode has ended and that conditions are becoming favorable for La Niña to develop.

Most of the Statistical and Ocean-Atmosphere coupled models, indicate additional anomalous cooling over the equatorial Pacific during the next 2-3 months. Some of the forecast models indicate a rapid transition to La Niña conditions during next few months.

Source : Press Information Bureau
Date : April 19, 2007