Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

Probability Distribution of Monthly Rainfall and the Surface of Assured Monthly Rainfall in Bangladesh

DR. D.N.R. Poul
CSO and Head
Agricultural Statistics Division, BRRI, Gazipur

Md.Shahjahan Kabir
SSO
Agricultural Statistics Division, BRRI, Gazipur

Abstract/ Executive Summary:

In Bangladesh most of the annual rainfall is concentrated during the months of May-October. Normal, Log-normal, Gamma, DP1 and DP2 probability distributions were fitted to the monthly rainfall data derived from 30 years’ daily rain fall data from May to October of 19 weather stations of Bangladesh Meteorological Department (BMD). The goodness of fit of the models was judged by Chi2–values. From the best fitted model obtained from monthly rainfall, total rainfall at 90% probability level had been estimated by solving its cumulative equation F(x,a,b..)=0.9 for x, where a,b.. are the estimates of the parameters of the selected probability distribution. These estimates for the minimum value of monthly rainfall were treated as the assured monthly rainfall of the specific weather station.

Keywords: Probability distribution, Rainfall, Probability distribution.

Location: All weather stations of Bangladesh

Start Date: 30-06-2001

End Date: 31-12-2003

Program Area: Socio-economic and Policy

Non-commodity: Rainfall

Objectives:

To determine the best fitted probability distribution for monthly rainfall and to construct the surface of assured monthly rainfall in Bangladesh.

Methodology:

Secondary data on rainfall, evapotranspiration of different weather stations of BWDB were used to find out relative water supply (RWS) which were used to determine the best fitted probability distribution for monthly rainfall and to construct the surface of assured monthly rainfall in Bangladesh.

Source of Information: BRRI Annual Report 2002-2003

Article Link (Online Journal):

Funding Source:

1. Government

Budget: 100000

Total Budget (Tk.) : 100000

Achievement/Findings:

After normal distribution, the second suitable distribution was the Log-normal, which fitted well in 47 cases and the third was DP2 distribution, which fitted well in 36 cases. In three stations, Gamma distribution fitted well to the monthly rainfall for 6 months and in 8 stations, Gamma distribution fitted well to the monthly rainfall of any 5 out of 6 months. No other single distribution was found to fit well to each of the monthly rainfall in any station. There were only 5 stations of which Log-normal distribution fitted in 2 stations, DP1 fitted in 2 and DP2 in one station for any 5 out of 6 months. Thus, Gamma distribution was found to be the best among the alternatives and hence adopted for graduating the probability distribution of monthly rainfall in Bangladesh. Using this probability distribution monthly rainfall surface at 90% probability level has been computed for the months of May – October for the whole country and the results are presented for the months June-September, the period, for which the rainfall is maximum (Maps 7-12). These maps can be used to predict the nearly assured monthly rainfall in May – October at any location of Bangladesh. The rainfall surface maps indicated that there would be no water problem in rain fed rice cultivation from June-September all over the country except few areas in the western part. However, supplemental irrigation may be needed in the western part due to low rainfall (<100 mm) during May for land preparation, planting and crop establishment for the Aus rice. Rice crops which are harvested in November and onwards, would need supplemental irrigation as these crops may face drought in reproductive phase due to low rainfall (<100 mm) during October.

Knowledge Management: Report/Proceedings