India

India: Long range forecast update for 2005 south-west monsoon season rainfall

Format
Other
Source
Posted
Originally published
INDIA METEOROLOGICAL DEPARTMENT

1. Background

On 20th April, IMD issued the first stage forecasts for the 2005 Southwest monsoon season rainfall (June-September) for the country as a whole. These forecasts were as follows:

IMD's operational Long Range Forecast for the 2005 South-west Monsoon season (June-September) is that the rainfall for the country as a whole is likely to be 98% of the Long Period Average (LPA) with a model error of ± 5 %.

The probabilistic model suggests a very high (75%) probability for the 2005 south-west monsoon season rainfall over the country as a whole to be near normal and above.

2. Second Stage Forecasts

The following forecasts are being released now:

a) Quantitative and probabilistic forecast update for the South-west Monsoon Season rainfall for the country as a whole using the 10-Parameter models.

b) Quantitative forecast for the July rainfall for the country as a whole using the 8-Parameter Power Regression Model with the model error ± 9%.

c) Quantitative forecasts for the South-west Monsoon Season(June-September) rainfall for the following four broad homogenous regions of India:

Northwest India -- Jammu and Kashmir, Himachal Pradesh, Punjab, Rajasthan, Haryana, Chandigarh, Delhi, Uttaranchal and Uttar Pradesh.

Northeast India -- Arunachal Pradesh, Meghalaya, Assam, Nagaland, Manipur, Mizoram, Tripura, Sikkim, West Bengal, Bihar and Jharkhand.

Central India -- Gujarat State, Madhya Pradesh, Chattisgarh, Maharashtra, Goa and Orissa.

South Peninsula -- Andhra Pradesh, Karnataka, Tamil Nadu, Kerala, Lakshadweep and Andaman and Nicobar Islands.

For this purpose, individual Multiple Regression Models with model error of ±8% have been used.

3. Onset and Advance of the Monsoon in 2005

Using an indigenously developed statistical model, IMD predicted that monsoon onset over Kerala would take place on 7 June with a model error of 3 days. This year, monsoon onset over Kerala was on 5 June, 4 days later than its normal date. By 7th June, monsoon covered the coastal Karnataka and Goa. However, there was a lull in further advancement of monsoon till 16 June. From 18 June, monsoon rapidly advanced along the west coast and by 21 June, monsoon covered Konkan and parts of Gujarat State. Monsoon advanced over Northeast India only on 16 June, almost two weeks behind its normal date. On 24 June, monsoon progressed rapidly into remaining parts of Maharashtra, Chattisgarh, Jharkhand, and parts of Madhya Pradesh. By 26 June, as monsoon became active, it further progressed over Uttar Pradesh, parts of Punjab and Haryana. Monsoon covered the whole country by 30 June, almost two weeks ahead of its schedule. During the period, 1960-2004, there were only 7 occasions (1960, 1961, 1975, 1980, 1994, 1996 and 1998) when monsoon covered the whole country in June itself.

Monsoon was sluggish during the first half of June. By the third week of June, monsoon became active, but only along the west coast. By 22 June, rainfall over the country as a whole was large deficient by 49%. However, during the last week of June, monsoon revived vigorously, which improved the rainfall situation over the country. By 30 June, cumulative rainfall deficiency over the country as a whole has significantly reduced to 15%. During the period 1 to 30 June, out of the 36 meteorological sub divisions, 19 sub divisions received excess /normal rainfall and 17 sub-divisions received deficient rainfall.

4. El Nino Conditions

Over the Pacific Ocean, SSTs have remained above normal but below the El Nino thresholds. Trade winds over the Pacific Ocean are stronger than normal. Southern Oscillation Index (SOI) value also showed substantial increase and it became positive. The latest El Nino predictions suggest higher probability for near neutral conditions during the next 3-4 months. However, the likely evolution of SST anomalies over the Pacific Ocean remains a matter of uncertainty. IMD is carefully monitoring the developments.

5. New Initiatives

Under a collaborative research programme with the Indian Institute of Science, Bangalore, a dynamical prediction system was installed at the National Climate Centre, Pune. For this purpose, the seasonal forecast model of Experimental Climate Prediction Center (ECPC), Scripps Institution of Oceanography, USA was adopted. The experimental predictions based on May sea surface temperature data suggest above normal rainfall over the country as a whole during the 2005 monsoon season (June to September).

IMD also continued its consistent efforts to further improve the statistical models. Several new statistical models were developed indigenously based on objective methods like ensemble linear regression, artificial neural network and projection pursuit regression.

In addition, IMD also consulted the experimental forecasts for the 2005 southwest monsoon rainfall prepared and supplied by Indian Institute of Tropical Meteorology, Pune, Indian Institute of Science, Bangalore, Space Applications Centre, Ahmedabad and the National Center for Medium Range Weather Forecasting, New Delhi.

6. Forecast Updates for 2005

(i) South-West Monsoon Season Rainfall

IMD's Long Range Forecast update for the 2005 South-West Monsoon Season (June-September) is that for the country as a whole the seasonal rainfall is likely to be 98 % of the Long Period Average (LPA) with a model error of ± 4%.

The probabilistic model suggests a very high (70%) probability for the 2005 south-west monsoon season rainfall over the country as a whole to be near normal and above.

ii) July rainfall

Rainfall in the month of July 2005 for the country as a whole is likely to be 97% of its LPA with a model error of ± 9 %.

iii) Rainfall over homogenous regions

Over the four broad homogenous regions of the country, rainfall for the 2005 South-West Monsoon Season is likely to be 97% of its LPA over North-West India, 95% of its LPA over North-East India, 102% of its LPA over Central India and 97% of its LPA over South Peninsula, all with a model error of ± 8 %.