Sunflower area and production variability in Pakistan: Opportunities and constraints
The edible oil imports bill rising from Rs. 77 million in 1969-70 to Rs. 3,900 million in 2002-03 has overburdened the economy of the country. Only 30% of the total needs are met through local production, while 70% are provided by import. Major share of the domestic production of edible oil comes from cottonseed and canola, 67 and 19.6%, respectively. The remaining 13.4% are contributed mainly by sunflower. Although it is a high oil, high yielding crop that gives high returns to the farmers, no
... to the farmers, no serious effort has been made to increase the local production of sunflower. Consequently, the sunflower acreage declined from 144,191 ha in 1998-99 to 107,717 ha in 2002-03 and the production from 194,544 to 128,531 t during the same period. The 1998-99 acreage was the maximum area under sunflower achieved. The big fluctuations in sunflower acreage and production are due to its price on the market. In the period of last 15 years, the sunflower acreage in Pakistan expended from 29,500 to 107,700 ha. The sunflower production rose at the annual rate of 9.9%, comprised of a 9.7% expansion in acreage and a minor improvement in productivity amounting to 0.16%. This increase was not sufficient to meet the requirements of the country. There is a big gap between the potential and actual yields of sunflowers. More than 70% of the potential have not been achieved yet. For this purpose the R 2 value was also calculated and, keeping in view the fluctuations in the time series data, second-degree equation was also measured. Logarithmic and exponential functions were also tested but the variability in the data measured by the R 2 value was best represented by seconddegree polynomial function. When the data seem to depart more or less widely from linearity in regression or time series analysis we must consider fitting some other curve instead of the straight line. The R 2 value was also improved with second-degree polynomial function for production from 43% to 58% showing a better fit of the trend line. The sum of the error terms was "0" for second-degree polynomial function but it gave a better fit due to a higher R 2 value. The higher b value for production portrays an increase in the productivity. The sum of squares for the estimated and observed values was 0. However, due to a low value of the coefficient of determination with linear trend and variation in the data, second-degree polynomial function (parabola) was estimated which gave a higher value of the coefficient of determination. With the use of * Corresponding author second-degree polynomial function the value of coefficient of determination increased from 50% to 60%.