A COMPARISON OF METHODS TO DETECT THE ONSET OF THE GROWING SEASON AND ITS TRENDS FOR SOME STATIONS IN THE SUDAN SAVANNA OF NORTHERN NIGERIA
A COMPARISON OF METHODS TO DETECT THE ONSET OF THE GROWING SEASON AND ITS TRENDS FOR SOME STATIONS IN THE SUDAN SAVANNA OF NORTHERN NIGERIA
No Thumbnail Available
Date
1995-10
Authors
OJONIGU, FRIDAY ATI
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
For rainfed agriculture, the major interest of the fanner is in the probability that, at certain
moments in time (i) the rains will become fairly continuous and enough to ensure sufficient
moisture in the soil at the time of planting and (ii) that this level will be maintained or even
increased as the season advances.
In the semi-arid tropics the start of the rains is seldomly abrupt. Rather, it is preceded by
a succession of isolated showers of uncertain intensity with intervening dry periods of varying
duration. Besides, what happens at the beginning of the rainy season to a certain extent
determines the characteristics of the season. The ability of effectively predicting the start of the
effectively productive rains would have great consequences for the total agricultural production
of the season.
Five methods for the determination of the onset date were selected. One is based on a
traditional technique; two on accumulated rainfall totals; and two on rainfall evapotranspiration
relationships. These were applied on data from eight stations in the Sudan Savanna of Nigeria.
The data span several years. The performance of these methods were -ested for each station. The
techniques were further tested for false starts.
The traditional "Ramadan" technique performed most poorly of all the methods. The
techniques based on accumulated rainfall totals performed diversely. Walter's method gave
early onsets and was closer in its poor performance to the traditional "Ramadan" method.
Sivakumar's method gave very late onsets, thereby shortening the growing season. The
techniques based on rainfall/ evapotranspiration relationship behaved closely to each other and
fell in between Walter's and Sivakumar's methods. They are, however, significantly affected
by false starts (up to 33% of the years).
To avoid the trap of false starts and shortening of the growing season, a combination of
Kowal's and Sivakumar's criteria (omitting the three consecutive rainy days) were used to
propose an alternative technique. This proved to work well for determining the onset date in the
study area.
Both the rainfall series and the onset series for Kano were tested for trends. While the
rainfall series showed a negative trend, the onset series did not indicate any statistically
significant trend. An abrupt change was discovered for the rainfall series around 1967. Also an
abrupt change was discovered for the onset series around 1987 for the first time. There is also
a strong inverse relationship between the rainfall series and the onset series. The nearing abrupt
change in the onset data makes the use of average onset dates questionable.
Due to uncertainty in the use of average onset dates, it is suggested that an advisory body
should be constituted. This body would be responsible for calculating the date of onset for any
year and diseminate such date to the fanners.
Description
A THESIS SUBMITTED TO THE POSTGRADUATE SCHOOL
AHMADU BELLO UNIVERSITY, ZARIA,
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR
AWARD OF THE DEGREE OF MASTER OF SCIENCE (M.Sc.)
IN GEOGRAPHY OCTOBER 1995
Keywords
COMPARISON,, METHODS,, DETECT,, ONSET,, GROWING,, SEASON,, TRENDS,, STATIONS,, SUDAN,, SAVANNA,, NORTHERN,, NIGERIA.