Yazar "Susam, Tekin" seçeneğine göre listele
Listeleniyor 1 - 4 / 4
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe COMPARISON OF GIS BASED INTERPOLATION METHODS IN ASSESSING OF SITE SPECIFIC PHOSPHORUS VARIABILITY ON THE APPLE ORCHARD(PARLAR SCIENTIFIC PUBLICATIONS (P S P), 2013) Karaman, M. Rustu; Horuz, Ayhan; Susam, Tekin; Er, Fatih; Tusat, EkremEvaluating of geostatistical approaches in monitoring of spatial variability of some chemical contaminants such as agricultural phosphorus (P) will provide valuable data for large agricultural areas. In this study, performance of varied GIS based geostatistical interpolation methods were tested in assessing of site specific P variability on the apple orchard. For this aim, soil samples were systematically collected from the agricultural apple area using the grid sampling system. The samples were taken at two depths (025 cm and 25-50 cm), the distance on the Y direction is 10 m and in the X direction is 20 m. The soil samples were prepared for analysis, and some physical and chemical analyses were made in the samples by routine methods. The data concerning with soil P levels were analyzed comparatively according to GIS based interpolation methods of Ordinary Kriging (OK), Simple Kriging (SK) and Universal Kriging (UK). The interpolation methods were also tested with varied semivariogram models of Spherical, Exponential and Gaussian. As a result of cross validations, the best optimal method was found to be interpolation method of UK (Universal RMSE, +/- 0.472) with semivariogram model of guassian for topsoil, whereas it was found to be interpolation method of SK (Simple RMSE, +/-0.323) with semivariogram model of exponential for topsoil. Predicted P values were significantly (p< 0.01) correlated with measured values for both topsoil (r = 0.993) and subsoil (r = 0.980), respectively. Soil P distribution maps were adequately performed by using selected kriging interpolation methods and suitable sernivariogram models. The results indicated that monitoring of site specific P variability on the apple orchard using these GIS based interpolation methods will help to create the effective schemes for agricultural chemical managements such as P fertilization resulting in optimal yield and quality with reduced environmental pollution.Öğe Evaluation of Geostatistical Mapping Strategies in Monitoring of Spatial Distributions of Iron and Zinc on a Calcareous Barley Field(Wfl Publ, 2010) Susam, Tekin; Karaman, M. Rüştü; Er, Fatih; İşeri, İsmailGeostatistical information on spatial distributions of chemical properties of agricultural soils is important for refining farm managements and precision farming. Site specific monitoring of field soils, that is one of the main steps of precision agriculture, provides more accurate information especially for balancing iron (Fe) and zinc (Zn) levels on the varied soil and plant conditions. In this study, spatial variability of Fe and Zn levels on a calcareous barley field under the barley plants were monitored by performing geostatistical mapping strategies. For this aim, soil samples were systematically taken from the study area at two depths (0-20 and 20-40 cm), on a grid system 10 m x 10 m intervals in the E-W and N-S directions. Descriptive statistics indicated that the coefficient of variation for Fe was low as compared for Zn in both subsoil layer samples. The CV values for Fe and Zn levels were 25% and 33% in topsoil and 23% and 29% in subsoil, respectively. Geostatistical analysis techniques were used in predicting the spatial structure of Fe and Zn levels of soil and plant. The spatial distribution maps were obtained by using Simple Kriging Method (SKM) with spherical semivariogram model for topsoil Fe, SKM with Gaussian semivariogram model for topsoil Zn. The Fe levels of topsoil and Zn levels of subsoil had moderate positional dependence, whereas Zn levels of topsoil and Fe levels of subsoil had weak dependence on the experimental field. The obtained range values were close to each other except for subsoil Fe and changed between 21 m and 15 m. The Fe and Zn levels of barley plants had moderate positional dependence. The results showed that site specific Fe and Zn levels of the field soil could be spatially varied within the small sampling points. The results have revealed that the data values of non-sampled points could be estimated by using SK method and suitable semivariogram model. Evaluation of geostatistical mapping strategies in monitoring of spatial distributions of Fe and Zn levels on a field will improve the decision support for field management practices in a more healthy and moderate way. It will also help to eliminate unequal micronutrient sources of the field soil, which is valuable for balanced crop nutrient consumption.Öğe Monitoring of Site Specific Fe and Zn Variability on the Apple Area Using the Gis Based Spatial Pattern Maps(AGRONOMSKI FAKULTET SVEUCILISTA U ZAGREBU, 2010) Susam, Tekin; Karaman, M. Ruestue; Er, Fatih; İşeri, İsmailGeostatistical approaches is the key issues of the modeling implementation in recent years, and it allows to figure out the spatially distribution of soil parameters. These methods will help to agricultural managements in a more healthy and moderate way, especially in precision farmings. In tilts study, spatial variability of soil chemical properties such as iron (Fe) and zinc (Zn) were examined on the agricultural apple area. Site specific variations of Fe and Zn on this area were predicted by performing GIS based spatial pattern maps. For this aim, the soil samples were systematically taken from the study area at two depths (0-25 and 25-50 cm). The grid system was used for locating the sample position. As a result of descriptive statistics, the coefficient of variation was lower for Fe levels when compared with Zn levels in both top and subsoil layers. The coefficient of variations for Fe and Zn levels were 24.7%, 42.5% for topsoil and 20.6%, 2.9.6% for subsoil, respectively. There was also a significant correlation (R-2=0.30, p<0.05) between topsoil Zn and subsoil Zn. Geostatistical analysis techniques were used for predicting the spatial structure of soil Fe and Zn levels. The spatial distribution maps were constructed by using Simple Kriging Method (SKM) with spherical semivariogram model for topsoil Fe, SKM with guassian semivariogram model for topsoil Zn. Based on the selected kriging method and semivariogram models, soil Fe and Zn levels were spatially varied within the study area. The maximum range was reached at 29 m for Fe level at the topsoil layer. The ranges were varied between 27 m and 23 m for top and subsoil Zn levels, respectively. The results have also revealed that soil chemical properties measured could he spatially varied within the small sampling points. For more accurate results, these geostatistical methods should be tested for varied conditions and spatial databases.Öğe Simulation of Organic Matter Variability on a Sugarbeet Field Using the Computer Based Geostatistical Methods(2009) Karaman, M. Rüştü; Susam, Tekin; Er, Fatih; Yaprak, Servet; Karkacıer, OsmanComputer based geostatistical methods can offer effective data analysis possibilities for agricultural areas by using vectorial data and their objective informations. These methods will help to detect the spatial changes on different locations of the large agricultural lands, which will lead to effective fertilization for optimal yield with reduced environmental pollution. In this study, topsoil (0-20 cm) and subsoil (20-40 cm) samples were taken from a sugar beet field by 20 × 20 m grids. Plant samples were also collected from the same plots. Some physical and chemical analyses for these samples were made by routine methods. According to derived variation coefficients, topsoil organic matter (OM) distribution was more than subsoil OM distribution. The highest C.V. value of 17.79% was found for topsoil OM. The data were analyzed comparatively according to kriging methods which are also used widely in geostatistic. Several interpolation methods (Ordinary, Simple and Universal) and semivariogram models (Spherical, Exponential and Gaussian) were tested in order to choose the suitable methods. Average standard deviations of values estimated by simple kriging interpolation method were less than average standard deviations (topsoil OM ± 0.48, N ± 0.37, subsoil OM ± 0.18) of measured values. The most suitable interpolation method was simple kriging method and exponantial semivariogram model for topsoil, whereas the best optimal interpolation method was simple kriging method and spherical semivariogram model for subsoil. The results also showed that these computer based geostatistical methods should be tested and calibrated for different experimental conditions and semivariogram models.
