کاربرد لاجیت ترتیبی تعمیم یافته در تعیین عوامل اجتماعی- اقتصادی مؤثر بر آلودگی آب‌های زیرزمینی (مطالعه موردی: آبشویی نیترات در کشت برنج)

Introduction: In today’s traditional agriculture and trade, it may not be possible to increase yields per hectare without applying some types of chemical fertilizers and pesticides. By producing 950 thousand tons of rice Mazandaran province supplies 42% of rice of the country. In the internal and external studies, it has been dealt with the study of entering nitrate into water technically. According to the destructive effects of nitrogen water-soluble on health and to the fact that Mazandaran province is the main producer of rice in Iran and applying the fertilizers containing nitrate is common in rice cultivation; the purpose of this study was to examine the effective factors on the entrance of nitrate into groundwater in cultivating rice in Sari. The aim of the present study is also to examine the effective factors on absorption pure nitrogen in the water used to cultivate rice.
Materials and Methods: The dependent variable in this study indicates the amount of pure nitrogen used water by i th farmer that has been divided into three categories. Calculating the dependent variable is such that the total amount of urea consumed by farmer is obtained by interview with farmers, the dependent variable of the amount of Nitrate pollution used groundwater was divided into three classes of low, medium and high. Considering such studies and the nature of dependent variable of the present study, that is as ordered discrete, the best model to respond the research objective is to use logit ordered pattern, but if the parallel regression assumption has not been considered, the advanced generalized ordered logit pattern should be used. In generalized ordered logit, the estimating parameters of independent variables that have rejected the assumption of parallel regression, cannot be same for different groups or levels. In other words, each level not only has separate intercept but also different coefficients. In order to select the samples, simple random sampling method has been used. Data used in this section has been collected by survey research method in 2015 for Sari. In order to determine the sample size, a pre-study has been conducted based on Cochran relation. The sample size of farmers’ was set at 98 farmers but finally 106 questionnaires were completed. Estimating the model and their results were conducted through Stata v.12 software.
Results and Discussion: According to the results, the dependent variable has been divided into three groups. About 43% of rice producers are the farms who used a moderate amount of nitrate into water, but 64% of them used medium and high amount of nitrate pollution that should be noticed. The results of parallel regression test for individual independent variables showed that among 17 independent variables, seven variables have been violated the condition of parallel regression. Therefore, the generalized ordered logit model was used. The variable income that has not been violated the assumption of parallel regression, has the coefficient of 0.83 in the first and second groups, that has been significant at the level of 10%. It means that, by increasing the level of farmers’ income and stability of other conditions, the possibility that the farmer will be in the group with more pollution will increase. As expected, the significant variable such as applying animal fertilizer, paying water expenses and agreement with omitting subsidy have negative and significant effect on the amount of nitrate used water. In other words, by increase in each variable, the possibility of placing farmers in the group of more pollution decreases. If farmers’ income increases as one unit, the possibility of placing the farmer in the group of low pollution decreases 0.2 unit and placing in the group of high pollution increases 0.17 unit. Being familiar with cultivating organic production will increases the possibility of placing the farmer in the group of low pollution in 0.55 unit, while the possibility of placing in the group with high pollution increase in less amount. On the other hand, the more positive attitude in rice cultivator towards supportive and public organization, the possibility of their placing in the group of low pollution of Nitrate decreases 0.13 units and the possibility of their placement in the group of high pollution on Nitrate increases 0.11 units.
Conclusions: Based on the obtained results and coefficient for every level of dependent variable, entrance of Nitrate in to groundwater in cultivating rice in Sari and interpreting the final effects of significant variables, the recommendations are offered for optimal utilization of fertilizers containing Nitrate and improving health of groundwater in cultivating rice as follows According to the factors effective on income and this variable that cultivating rice could not afford life express of 77% of the surveyed people, public supportive, private and insurance organizations should pay specific attention to this significant and important section. Revising supportive policies of the supportive public and private organizations for rice cultivators can be effective so that the policy making should have been lead into facilitate solving production and improving the farmers’ income than subsidiary supports of fertilizers. Revising in owner share and share on transaction under the supervision organs of agriculture are recommended. Of course income support and facilitating the sales of owner farmer can avoid renting the farm. Applying natural fertilizers and being familiar with cultivating organic and healthy products, have placed the rice cultivator in the level of minimum usage of chemical fertilizer as well as nitrate entrance into groundwater. Therefore, promoting, advertising and encouraging farmers to cultivate organic can be a strategy accompanied with financial and security incentive and finally, it is recommended to design and perform efficient policy making of water market in Mazandaran province to approach the real price of water.