عنوان مقاله [English]
Introduction: The importance and the presence of spatial variability in soil properties is inevitable, however, the understanding of causes and sources of the variability is not complete. Spatial variation of soil attributes can affect the quality and quantity of plants. Investigation of the soil variability at the small scale can be evaluated by classic statistics and geospatial statistics. The present study was conducted to investigate the spatial variability of yield characteristics of rose (Rosa Damasceneea Mill) and soil characteristics in two main cultivated fields of rose (Negar- Golzar) with different climatic and topographic characteristics located in Bardsir city, Kerman Province.
Materials and Methods: In order to achieve the objectives of the present study, 100 soil and plant samples were collected from each farm. The soil samples were taken from 0 to 25 cm depth and analyzed. The measured soil properties at each location were including fragment, clay, silt, sand, and organic matter contents, CEC, calcium carbonate equivalent, EC, pH, total nitrogen, available phosphorus, and available potassium. Moreover, some plant characteristics (yield, plant height, and plant crown diameter) were measured at each point. Then, maps of soil properties and plant induces were prepared using Geoeas, Variowin, and surfer software. Descriptive statistics were applied using Statistica software (version 20). Kolmogorov-Smirnov test was also used to test the tolerance of variables distribution.
Results and Discussion: The results of Kolmogorov-Smirnov test showed that all characteristics of the plant and soil in both farms follow the normal distribution. Statistical analysis showed that coefficient of variation of soil properties was as follows: total nitrogen (54.47%) and pH (3.16%) in Negar farm, and EC (46.09%) and pH (35.3%) in Golzar farm. The variability of nutrients in both farms had similar trends, so that total nitrogen, phosphorus and potassium have the highest to lowest coefficients of variation, respectively. Analysis of variograms indicated that all of the variables in both fields have a strong and moderate spatial variability. Ranges for variograms were from 122.16m (for yield) to 218.46 m (for silt) in Negar farm and from 115.1m (for available K) to 228 m for (total nitrogen) in Golzar unit. The distribution conditions and spatial variations of the soil properties in the study area were not uniform due to variation of the range of the variograms. The results also showed that the yield characteristics of the rose with some soil characteristics have a closer spatial relationship. About this, in the Negar farm, the range of the rose flower yield was close to the clay, available potassium and calcium carbonate contents. In the Golzar farm, the range of rose flower yield was close to the range of clay, silt, fragments and available phosphorus contents. The spatial correlation ratio showed that all plant characteristics including plant yield, plant height and plant diameter had a strong spatial correlation in the Golzar farm, and all characteristics of the soil were in the medium spatial correlation. Also, in the Negar farm, the product yield characteristics were in a strong spatial correlation class, and all other characteristics were in the medium spatial correlation. Kriging maps showed that soil characteristics and product yield in the study area had spatial distribution. The similarity of the spatial distribution pattern of some variables was one of the important features that these maps showed.
Conclusion: The results of this study showed the characteristics of plant yield and soil characteristics have a moderate to strong spatial dependency even in small scales. Kriging maps illustrated that the pattern and distribution of soil properties even within a farm can be varied. However, the spatial pattern of some soil characteristics such as organic matter and total nitrogen with the spatial pattern of plant characteristics and the dimensions of the farms showed conformity. This indicates that the variability of these characteristics is mainly under the management of farmers, and in order to optimize the use of nutrients, inputs should be re-evaluated by farm managers. In general, the results of this study indicated geostatistical method can be used to recognize of control factors of plant production and use its information in order to improve management.