| dc.description.abstract | The purpose of this study was to assess the relationship between weather parameters on cotton
yield. Weather is one of the important factors that affect crop growth and yields. Agricultural
activities in Tanzania are rainfed and weather-dependent practices so fluctuations in weather
parameters affect many of the crop yields. Cotton is Tanzania’s largest export cash crop after
coffee. It has been observed that cotton yields are generally decreasing. Because of that, there is
a need to examine the relationships between weather parameters and cotton yield over Mwanza
region in Tanzania.
At total of 15 years of data, of both weather parameters and cotton yield were collected from
Ukiriguru meteorological station and Ukiriguru agricultural research center respectively.
Weather parameters and the cotton yield were correlated and regressed based on different
phenological phases with the aid of a SYSTAT program to obtain regression equations that can
predict yield based on different phenological phases of the crop.
The results obtained from correlation analysis based on different phenological phases were, from
planting to leaf development rainfall and minimum temperature were found to be statistically
significant with correlation coefficient of 0.3285 and -0.5747 respectively. During the flowering
phase only the maximum temperature was found to be statistically significant with correlation
coefficient of 0.3802. From flowering to ripening rainfall was found to be statistically significant
with correlation coefficient of 0.3171. Lastly from planting to ripening rainfall was found to be
statistically significant with correlation coefficient of 0.4124.
These weather parameters and the cotton yield were regressed using a backward stepwise
regression technique with the aid of SYSTAT program at different phenological phases; it was
therefore found out that the variation in cotton yield is influenced by weather parameters.
Different regression models were developed in three different phenological phases, namely from
planting to leaf development, flowering to ripening and from planting to ripening. These
regression models can help to predict yield at different phases of the crop.
Therefore, this study has found out that weather parameters can affect the yield differently at
different phases during its growth, this imply that every stage of the crop development is crucial
in determining the crop yield.
incorporate long period data (of at least 30 years).
I also recommend that further studies should be carried out to determine the influence of other
weather parameters and the non weather parameters such as evaporation, wind speed, soil
temperatures, type of the cotton (Cotton varieties), and type of soil and fertilizer application. | en |
