Assessment of Summe Savory (satureja hortensis L.) Biomass by Easily-Attainable Soil Parameters and Artificial Network

Abstract

Introduction: One of the most important requirements in planning production and processing of medicinal plants in order to obtain high yield and high-quality is the initial assessment of the physical and chemical properties of soil, which reduces the production cost by avoiding the use of unnecessary soil analysis. Summer savory (Satureja hortensis L.) is one the most widely used medicinal plants that quality index of plant is related to the quantity and the constituent of its essential oil content. Understanding the relations between the quantity and quality of medicinal plants with the very physical and chemical properties of soil is very complex and the estimation of parameters changes of medicinal plants affect by soil quality characteristics is more difficult. Today, with the arrival of multivariable regression models and artificial lattice models in the research, many complex relationships found in nature is understandable. Hence the need for estimation the biomass yield of savory using fast, cheap and with acceptable accuracy is feeling. Materials and Methods: The present study was performed at the Agricultural Research Station Neyshabur as pot experiment based on a completely randomized design with three replications. Around 53 soil samples were collected from different parts of Neyshabur city, and soil texture, organic matter, pH, salinity, phosphorus, potassium, nitrogen and carbon content were selected as the easily available parameters. Before planting the parameters were measured in laboratory. Approximately 90 days after planting seeds in pots containing soil samples, the sampling of plants was done based on the treatments. For drying, samples were placed for 24 hours in an oven at 40 °C. Finally, the relationship between the biomass yield and easily available soil parameters was determined using artificial neural network by Matlab7.9 software. Results and Discussion: The results showed that soil variability, is a key element in the management of valuable information on soil properties within a field and valuable information on soil properties within a field nature puts at our disposal. In yield modeling with 10 parameters for 53 soil samples, the best makeup hidden layer with Levenberg-Marquardt algorithm training as a hidden layer, 58 neurons, logsig threshold function for hidden layer and Tansig for the output layer were selected. High values of R2 and low levels of RMSE mentioned the proximity of the forecast data with measurement data and high accuracy of the model in summer savory biomass yield estimation. To obtain the most sensitive parameters, the sensitivity analysis was calculated using no-sensitive coefficient. So that, if the coefficient of a sensitive parameter is more than 1.0, the mentioned parameter, is one of the critical parameters of model. Accordingly, the parameters of organic carbon, nitrogen, phosphorus, organic matter, potassium, pH, salinity, clay, silt and sand respectively were selected as the most sensitive parameters. The addition of input parameters increases the value of R2 and reduces the RMSE during training, validation and test stages. This represents an increasing in the accuracy of model in estimation of biomass yield via increasing the input parameters. Models 1(soil texture) and 2(carbon) are not enough strong for biomass yield estimation. With increasing the experiment from 1 to 2, the potency of the neural network model 3(soil texture + carbon) significantly increased. Thus with an overview, the model No. 3 suggested as an improved model because with the minimum number of imputes produced equal output comparing the models with more inputs. Conclusions: Based on the obtained results, it seems that with the improvement of artificial neural network models and determining appropriate parameters, results to understanding the soil factors involved in the formation of savory plant biomass and better planning. Till leads to a cheaper and better product. Also, results showed that the artificial neural network has high accuracy in estimating the biomass plant Summer Savory. So that, the 80% of yield variability of the study area, presents by using the data of 10 readily available properties of the soil. Yield biomass of savory, largely depends on the soil texture, organic matter, carbon and the minerals of the soil. Since, this study is the first work to estimate the biomass of medicinal plants using artificial neural network, therefore recommended to use this method to estimate the yield and essential oil of other medicinal plants.

Keywords

• Summer savory
• Artificial neural network
• Easily accessible soil parameters