Effect of Soil Characteristics on Yield and Preparing the Crop Requirements Table of Oranges in Selected Areas of Iran

Iran is one of the most important countries in citrus (oranges) production. Citrus fruits are grown in different soils with a wide range of physical, chemical and fertility properties in the country, although some restrictions in the cultivated lands cause yield loss. In this regard, the present study was conducted to investigate the effect of physical, chemical and soil fertility characteristics on citrus yield in important areas under cultivation, the regression relationships of characteristics with yield, and the rating of soil and land parameters.

The 138 oranges orchards (118 orchards for rating and 20 orchards for validation) were selected in Fars, Mazanderan, Kerman and Guilan provinces. In each garden, a questionnaire was completed, a soil pedon was studied and soil samples were taken to carry out the appropriate physicochemical analyses. The selected soil and land characteristics were soil salinity (EC), exchangeable sodium percentage (ESP), pH, gypsum content, soil calcium carbonate (TNV), organic carbon (OC), clay, sand, silt, gravel, and soil available phosphorus and potassium contents. From the whole obtained data, 20 data were considered for validation purpose and the remaining data were used for modeling based on stepwise multivariate and simple regression methods. In these equations, the relationship between yield, as dependent variable, with soil and land characteristics, as independent variables, was investigated. Finally, land characteristics rating was obtained by the FAO method and the proposed crop requirements table was evaluated using the validation dataset.

The results of descriptive statistics analysis showed that the variance values for available potassium, sand, clay, gravel and TNV were high and for pH and OC and gypsum were negligible. Therefore, most soil properties have a wide range of variation which could be related to the fact that oranges are grown in a wide range of soil types. The value of TNV varied between 10 and 33.3%. The presence of carbonate in soil reduces the availability of macro- and micronutrient elements in direct and indirect manners. The average of EC in the studied orchards was 5.4 dS.m-1. Minimum, maximum and average of ESP were 1.7, 28 and 10.7, respectively. The lowest and highest salinity and sodicity were observed in Mazandaran and Kerman soils, respectively. Maximum, minimum and average percentage of gypsum were 12, 0.36 and 3.54%, respectively. The highest amount of gypsum was observed in Bam and Shahdad regions of Kerman province and the lowest gypsum content was observed in Mazandaran and Guilan provinces. The soil pH varied from 6.63 to 8.8 with the average of 7.8. The soil OC values were between 0.05 and 3.53% and its average was 0.89%, showing the fact that the most studied soils were poor in organic matters. The average of soil available phosphorus and potassium in the studied orchards for citrus was less than the critical level. The average, minimum and maximum of available potassium were 224, 100 and 360 mg.kg-1, respectively. The mean, minimum and maximum amounts of available phosphorus were 21.6, 8 and 45.9 mg.kg-1, respectively. According to the multivariate regression model, among soil properties, EC, ESP, TNV, gypsum, gravel, available phosphorus and potassium were selected by the model. The determination coefficient of the model was 0.95, indicating that these properties have the greatest effect on citrus yield. Simple regression equations demonstrated that TNV, gypsum, EC, ESP, sand, clay, gravel, available potassium and phosphorous had the highest correlation (R2 > 0.6); and soil OC and pH had the lowest correlation (R2<0.2) with yield. The equations also revealed that soil EC, ESP, gypsum, TNV and gravel percentage had the greatest effect in yield loss, and soil organic carbon, absorbed phosphorus and potassium had the greatest effect on increasing citrus yield. As stated in equations, reported permissible and critical thresholds for effective soil properties on citrus yield, were 2.4 dS.m-1 for EC, 5 for ESP, 1.5% for gypsum, 20% for TNV, 22 mg.kg-1 for available phosphorus, 280 mg.kg-1 for available potassium, 110 cm for soil depth, and >2 m for groundwater level. Finally, evaluating the proposed crop requirements table with validation dataset fitted between citrus yield and soil index, resulted in the determination coefficient value of 0.79, denoting the acceptable accuracy of proposed table.

Overall results showed that the main land limiting characteristics for orange production were soil salinity and sodicity, high amount of soil calcium carbonate and gypsum. Among unsuitable physical and fertility properties of soil, salinity and sodicity are the most effective factors affecting yield reduction. Consequently, proper management practices such as introducing cultivars compatible with these soil conditions, soil remediation and leaching operations to reduce soil salinity and sodicity are necessary. Furthermore, in most areas under orange cultivation such as Fars and Kerman provinces, the soil calcium carbonate content is more than the critical level for plant growth. In addition, the averages of soil available phosphorus and potassium were less than the critical levels, which should be considered for nutrient management of orchards. The proposed table of crop requirements seems to be accurate enough to conduct land suitability studies for orange varieties, especially cultivars grown in the north and south of the country.