Milad Alizadeh Badresh; Farhad Hosseinali
Abstract
Extended Abstract
Introduction
Cultivation Pattern is a roadmap that shows which, how much, when, and where crops should be cultivated given the constraints and available resources. Cultivation pattern program determines appropriate crop types in accordance with the climatic condition of the province ...
Read More
Extended Abstract
Introduction
Cultivation Pattern is a roadmap that shows which, how much, when, and where crops should be cultivated given the constraints and available resources. Cultivation pattern program determines appropriate crop types in accordance with the climatic condition of the province and thus ensures the sustainability of agricultural products, food security, and optimal utilization of resources, capabilities and potentials of each region. Review of the related literature indicates that AHP and TOPSIS methods are among the most widely used methods in decision making and prioritization. Moreover, previous studies have shown that AHP method is suitable for qualitative data and TOPSIS method is suitable for quantitative data, whereas both quantitative and qualitative factors are involved in determining the cultivation pattern. Therefore, the present study has utilized a larger number of criteria (nine criteria), and combined AHP and TOPSIS models in an attempt to make use of their strengths and avoid their weaknesses. Linear programming was also used with four scenarios. In one of the scenarios, two lowest ranking crops in TOPSIS method were eliminated. The present study has innovatively utilized these models and a larger number of criteria simultaneously to determine the cultivation pattern. It also has precisely identified the appropriate crop for each plot of land using SWOT tables.
Materials & Methods
The case study was located in Qeyghaj plain in west Azerbaijan province. In accordance with the geographical location and climate, wheat, barley, alfalfa, sugar beet, rapeseed, potato, maize and fodder corn are mostly cultivated in the area which have been considered as alternatives for cultivation in each plot of the present study. The present study has begun with evaluating the slope and aspect of each cultivation plot. Then, crops are ranked and optimal crops are selected based on various criteria and using a combination of AHP and TOPSIS models and different decision matrices. Afterward, the maximum and minimum appropriate volume of crop production is determined using linear programming in accordance with the maximum profit. Finally, the most suitable crops for each land parcel are determined using SWOT tables.
The present study has proposed a multicriteria decision model which includes the strong points of AHP and TOPSIS models and avoids their weaknesses. In order words, relative weights were obtained from AHP pairwise comparisons and also the compatibility index was evaluated using AHP model while crop alternatives were ranked using TOPSIS model. The hierarchical structure included the goal, nine criteria used to evaluate the strategies and eight strategies (options).
Matrices used in pairwise comparisons were all obtained from experts' opinions. These include comparisons made to determine the weight of each criteria to be used in the TOPSIS model, as well as pairwise comparisons made between options which could not be quantitatively compared. Then, the general structure of the hierarchical model was developed in Superdecision software and the final weights, compatibility index of each matrix and quantity of each product were obtained based on each of the indicators. The values were then entered into the TOPSIS model and used to rank the crops, compare different options and select the best crop.
Results and Discussions
In the first step, a slope map was produced for the study area using digital elevation model based on which an aspect map was also produced. In accordance with these maps, the physiological suitability of the study area for the cultivation of eight crop types was evaluated. Results indicate that the study area is physiologically very suitable for cultivation of alfalfa, suitable for wheat, barley and canola and fairly suitable for the other four remaining crops.
Then, pairs were compared in hierarchical analysis using expert opinions and the weights of criteria and crops were obtained. Then, weights were assigned to each alternative (crops) and decision criterion (nine selected criteria) using the TOPSIS model and more appropriate products were selected. A decision matrix was first created in TOPSIS. Some criteria such as economic index were initialized directly in accordance with the available quantitative values whilst the values of some other criteria (such as temperature whose quantitative values cannot be obtained) were initialized using the results of AHP.
In the next steps, a weighted normalized matrix was developed and positive and negative ideals were found. The distance between positive and negative ideals was calculated and then the ideal solutions were obtained. Finally, the score obtained by each alternative or similarity index was calculated. The closer similarity index is to one, the superior that alternative will be. Linear programming is a method in mathematics that finds the minimum or maximum value of a linear function on a polygon. The present study seeks to reach maximum profit under various restrictions such as water restriction, restrictions on area under cultivation and maximum and minimum amount of cultivated crop. Water restriction included all surface and subsurface resources for crop cultivation. Crop coefficients were defined as the need for crop irrigation. Water constraints included the constraints assigned to allocated water in spring, summer, autumn and total amount of allocated water.
Three scenarios were developed with or without the previously mentioned constraints. Then the goal function was changed in accordance with the MOTAD method and another scenario was developed. The scenarios are explained as follows:
Scenario 1 (Without any restrictions on the minimum and maximum crop yield): In this case, the goal was reaching the maximum profit and the restriction included the lowest amount of water consumption, regardless of the requirements in the study area. In this scenario, variables x1 (wheat), x5 (Canola) and x8 (fodder corn) were included in the cultivation pattern. Consequently, farmers' income was maximized and the amount of water consumption was reduced. However, obtained results were not acceptable in accordance with the regional and national policies since cultivation of most crop types will thus be stopped.
Scenario 2 (locally acceptable size and local farming customs and the restrictions assigned by the agriculture office): the present scenario seeks to maximize profit, satisfy requirements of the area and achieve the goals of the agriculture office. All crops are included in the cultivation pattern. Therefore, minimum and maximum cultivation restrictions have been used in addition to water and land restrictions.
Scenario 3 (not cultivating some water-intensive crops): As previously mentioned, Poldasht agriculture office has introduced reduced cultivation of some low yielding crops or even stopping the cultivation of such crop types as one of its main goals. Corn and potato are highly water -intensive with a low yield in the study area and thus gain one of the lowest ranks. Therefore, potato and corn were removed to determine the cultivation pattern of the region in their absence.
Scenario 4 (MOTAD approach): MOTAD is a linear programming approach aiming to maximize the profit whose objective function equals the sum of deviations between total gross income and the expected income based on the average gross income of the sample. Linear programming with MOTAD requires having access to income gained from each crop type in previous years. Restrictions such as fund and manpower restriction must also be considered. The statistical period used in MOTAD approach starts in 2011 and lasts till 2016.
Income values in MOTAD approach lead to a constraint relation. Just as the previous scenarios, water and land constraints are considered in this approach and fertilizers and pesticides restrictions have not been taken into account.
Conclusions
Based on the collected information, available parameters, SWOT analytical model and tables developed for each field, a suitable crop was selected for each farm (parcel). Accordingly, 112.3 hectares was identified as suitable for the cultivation of wheat, 59.9 hectares for barley, 32.1 hectares for alfalfa, 37.6 hectares for sugar beet, 85.7 hectares for Canola, 15.5 hectares for potato, 13.2 hectares for Maize and 63.7 hectares for fodder corn. In this case, the resulting profit equaled 23, 503,410,000 Rials and the water consumption equaled 2,542,293.8 cubic meters which shows 2,052,120,000 Rials increase in profit and 90,770.6 cubic meters decrease in water consumption as compared to the present cultivation pattern.
Comparing the profit and water consumption in each of the five models and the current cultivation pattern, it can be concluded that the pattern obtained from the SWOT analytical model is more feasible since it includes various parameters and particularly farmers' opinions.
Amir Hossein Kazem; Farhad Hosseinali; Ali Asghar Ale-sheikh
Abstract
Modeling urban growth and land use changes are an integral part of planning for sustainable development. The present research intends to model the urban growth and development for Tehran metropolis from the aspect of timeand spatial distribution. To this end, land-use maps for the years 1988, 2002 and ...
Read More
Modeling urban growth and land use changes are an integral part of planning for sustainable development. The present research intends to model the urban growth and development for Tehran metropolis from the aspect of timeand spatial distribution. To this end, land-use maps for the years 1988, 2002 and 2013 were categorized with the object-based approach using Landsat satellite time series images. In the next step, using the logistic regression model, the effect of independent variables in relation to urban growth including 14 variables in the form of two groups of environmental-natural and socio-economic variables during the period of 1988 to 2002 was calculated as the coefficient in the regression equation, and the potential map of urban expansion was produced. The evaluation of the logistic regression function using two Pseudo R2 and ROC indexes with values of 0.32 and 0.89 showed good regression fit and proper description capability. Subsequently, the area of change for the expected year was quantitatively predicted using Markov chain analysis.Finally, by using the outputs of the two models of logistic regression and Markov chain analysis and using the Cellular Automata Model, urban growth was modeled for the year 2013, comparison of which with the 2013 classified image, shows that the used model with a 93% relative accuracy for the estimated area and a Kappa coefficient of 0.87 has been a successful model. Accordingly, the same model was used to estimate the urban growth in 2025,using images from the years of 2002 and 2013.
