At present, an objective method of reporting of area and production under flowers is not available. Although National Horticulture Board (NHB) publishes area and production figures, these are not based on any statistical and scientific procedure. The already existing crop-cutting experiment approach being followed in case of other crops would be cost-prohibitive for developing estimates of area and production of flowers. This is due to multi-pickings involved in case of flowers. In this context, the National Statistical Commission has recommended the need to develop a suitable sampling methodology for estimation of area and production of flowers on the basis of market arrivals. With the growing demand for the need of building reliable estimates of area and production of flowers, the present study has been planned to meet the requirement of reliable estimates on area and production of flowers based on the statistical methods. Therefore Indian Agricultural Statistics Research Institute conducted a study entitled “Pilot sample survey to develop sampling methodology for estimation of area, production and productivity of important flowers on the basis of market arrivals” funded bythe Ministry of Statistics and Programme Implementation, Central Statistical Organization with the objective of developing precise estimates of the production of important flowers on the basis of market arrivals in Delhi and the estimates of area under important flowers on the basis of flower growing villages in Delhi.
As a part of this study, a sample survey was carried out in the three flower mandis of Delhi namely, Khari Baoli mandi, Hanuman Mandir mandi and Mehrauli mandi and in the randomly selected flower growing villages of Delhi during September 2003 to August 2004 to collect the data on market arrivals of important flowers in the flower mandis and data on area and production from the randomly selected villages. Keeping in view the trend of flower arrivals in the mandis, the entire survey period was divided into three periods viz. P-I: September-December 2003, P-II: January-April 2004 and P-III: May-August 2004. Cut flowers of Rose, Gladiolus, Chrysanthemum, Tube-rose, Carnation and other seasonal flowers are mainly traded in the Hanuman Mandir Mandi. In Fatehpuri and Mehrauli Mandi, trading is done in loose flowers of Marigold, Rose, Margaret and Jaffrey etc.
The sampling design for estimation of production on the basis of market arrivals was one of stratified random sampling in each flower mandi. The commission agents/mashakhors in the mandis of Delhi comprise the first stratum while self-selling farmers the second stratum. The sampling design for estimation of area under floriculture in flower growing villages of Delhi was one of stratified single stage random sampling with flower growing villages as the sampling units. For estimating productivity of important flowers on the basis of village survey, the sampling design adopted was stratified two stage random sampling with villages as first stage sampling units and farmers growing flowers as the second stage sampling units.
For estimating area and productivity of important flowers in the flower growing villages of Delhi, 15 flower growing villages were selected randomly. The flower growing villages are classified in to three strata. The villages having area under floriculture up to 5 ha comprise first stratum while the villages having area from 5 to 10 ha comprise the second stratum and more than 10 ha, the third stratum. Area under flowers in the selected villages was recorded on complete enumeration basis. For estimating production of important flowers, all the flower growing farmers were selected in the randomly selected villages having 15 or less flower growing farmers, while 15 flower growing farmers were selected randomly in the villages having more than 15 flower growing farmers. Data on area, production and disposal practices were collected by inquiry.
Overall estimates of market arrival in Khari Baoli flower mandi from Delhi and outside Delhi were 13331.613 MT with 2.73% S E and 19914.467 MT with 1.93% S E respectively. Of the total market arrival of flowers in this mandi, two fifth was from Delhi. Period-wise total market arrival of cut flowers (in lakh) from Delhi in Hanuman Mandir flower mandi was estimated as 354.35129, 231.60071 and 84.73619 while from outside Delhi, the same was of the order of 809.47600, 528.28154 and 1043.04484. Total market arrivals from Delhi and outside Delhi were 670.68820 with 1.53% S E and 2380.80237 with 0.74% S E respectively. The results clearly indicate that of the total market arrival, both for period I and period II, 30% was from Delhi while for the third period it was only 8%. Pooled over all the three periods, the market arrival from Delhi and outside Delhi was 22% and 78% respectively. In Mehrauli flower mandi, the pooled estimate of Jaffrey from Delhi was observed highest 595.594 MT with 5.41% S E followed by Marigold 307.052 MT with 4.41% S E, Rose 175.788 MT with 1.37% S E and Margaret 152.281 MT with 6.72% S E. The estimate of other flowers (Dodhi, Raibel, Caner and Dhela) from Delhi was observed to be 0.854 MT with 10.10% S E while from outside Delhi it was 231.496 MT with 1.22% S E. It can be seen that both for period I and II, two fifth of the market arrival were from Delhi whereas it was only one tenth for period III. Pooled over the three periods, the market arrival from Delhi and outside was 17% and 83%.
Total market arrivals of loose flowers from Delhi as well as outside Delhi were found to be 14570.910 MT with 2.51% S E and 25829.580 MT with 1.50 % S E respectively. The corresponding figures for cut flowers were 670.68820 lakh with 1.535 S E and 2380.80237 lakh with 0.74% S E. The percentage of market arrival from Delhi and outside Delhi respectively was of the order of 36% and 64%. The corresponding figures in case of cut flowers were 22% and 78%.
Overall estimate of area in the flower growing villages of Delhi under cut flowers was 442.59 ha with the figure 259.09 ha in stratum III and 183.50 ha in stratum II. The estimated area under loose flowers was substantially higher in stratum III (2098.26 ha) than the area reported in stratum I and II. A total area of 2583.28 ha was estimated under loose flowers in Delhi. The total estimated area under both flowers was 3025.87 ha in Delhi. As a whole, 85.37% of the area under loose flowers while 14.63% under cut flowers is estimated in Delhi.
The period-wise estimates of the production of cut flowers in lakh numbers were 359.11495 with 1.98% SE, 257.38033 with 5.84% SE and 116.91642 with 17.97% SE in the three periods respectively. The overall production of cut flowers in was of the tune of 733.41170 lakh with 3.54% SE. Estimation production of loose flowers was reported substantially higher in stratum III. Estimated production of loose flowers was observed maximum 8277.3 MT with 4.31% SE in period II of stratum III. Pooled over all periods, these figures were 2992.1 MT with 6.80% SE, 1159.8 MT with 4.95% SE and the highest 13540.7 MT with 8.18% SE for the three strata respectively. The period wise pooled estimates of production of loose flowers was significantly higher in period II of the order of 10163.7 MT with 3.73% SE followed by 6272.4 MT with 7.47% SE in period I and 1292.5 MT with 10.62% SE in period III. The overall estimated production of loose flowers in Delhi was of the tune of 17728.7 MT with 4.09% SE.
Period wise and stratum wise estimates of productivity of loose and cut flowers in the flower growing villages of Delhi were also obtained. The period wise estimates of productivity of loose flowers were 7652, 10654 and 1596 kg/ha respectively. Overall estimated productivity of loose flowers in Delhi was 6863 kg/ha. The period wise estimates of cut flowers in numbers were 283035, 160172 and 73371 per ha. The over all estimated numbers of cut flowers in Delhi was found to be 148034 per ha.
The results of the above study clearly reveal that the percentage standard errors of the estimates of different kind of flowers traded in the three flower mandis of Delhi ranged from 0.94 to 7.75. On the basis of survey conducted in the flower growing villages of Delhi, estimated production of loose and cut flowers was 17728.7 MT with 4.09% SE and 733.41170 lakh with 3.54% SE respectively. However, the flower wise estimates of production on the basis of village survey are not given in the report as the area figures under each kind of flowers in the flower growing villages were not available. A separate study can be taken for developing sampling methodology for estimation of area and production of important flowers grown in the flower growing villages of Delhi on the basis of complete enumeration. The estimates so generated will form an important data base which can be utilized by the organizations direct connected with the floriculture sector.
As already mentioned that area and production figures under floriculture in different states of India published by other agencies are not based on any scientific/statistical procedure because they do not followed any statistical methodology for estimating these figures. On the basis of this study conducted in the flower growing villages and flower mandis of Delhi, both loose and cut flowers are being produced in Delhi and the estimates of the same have been provided in this report. In this study, statistical methodology adopted for estimating area and production figures of the important flowers clearly indicates its appropriateness and the results obtained from this study are more reliable and precise.
It may be pointed out that the above study conducted for estimating production of important flowers on the basis of market arrivals in Delhi is first of its kind and was possible because most of the flower produced in the villages of Delhi arrived in the flower markets situated in Delhi. Although the results obtained by this study are encouraging but these results pertain to union territory of Delhi. The methodology developed by us for estimating production of important flowers on the basis of market arrival needs to be tested in some other states before it can be recommended for adoption.
