Study On The Model For Traits And Dry Matter Partitioning And Yield Prediction In Salvia Miltiorrhiza Bunge

Salvia miltiorrhiza Bunge is an important traditional Chinese medicine, it has a long history of medical in our country. As the production of wild Salvia cannot meet the market demand, the planted area had expanded each year. In order to meet the requirements of modern medicine, make cultivated Salvia good quality, high yield and sustainable production, and explore rhythm of growth and dry matter partitioning in Salvia under different physiological and ecological conditions, it has great significance that establish and study the model for traits and dry matter partitioning and yield prediction in Salvia. The study includes four parts.1. Study on the Growth Rhythm Model of the Root Traits of Salvia Miltiorrhiza based on logistic equationThis paper proposed the growth models of root diameter, root length and root fresh weight by logistic equation. The data of2011and2012were used to fit and validate the models respectively.1:lplot and root mean squared error (RMSE) was used to compare simulated values with observed values. By obtaining the first, second and third derivative and inflection points of logistic equation, the growth period of Salvia miltiorrhiza was divided into the early growth stage, fast-growing stage and the late growth stage, and we analysis the relation of the growth rate of root fresh weight with the planting density and nitrogenous fertilizer respectively and the growing Salvia miltiorrhiza. The determined coefficient (R2) of the data of2011were higher than0.96and those models reached significant level (P<0.01). The determined coefficient (R2) of simulated values and observed values of2012were higher than0.95. The results show that there are highly consistent between the growth of root trait and logistic curve fitting. The growth rate of root fresh weight under low density treatment is faster than under the high density treatment in the middle and late growth; The moderate nitrogenous fertilizer (100kg/ha) make the growth rate faster and the growth time longer, and it is the very important time when reached fast-growing stage of Salvia miltiorrhiza root.2. Study on the growth rhythm Model of the stem fresh weight, leaf area and leaf fresh weight based on the Gauss function This paper proposed the growth models of stem fresh weight, leaf fresh weight and leaf area by the Gauss function, using accumulated product of Thermal Effectiveness and PAR (TEP) and effective accumulative temperature (GOD) as independent variables respectively. The data of2011and2012were used to fit and validate the models respectively.1:1plot, fit residuals and root mean squared error (RMSE) was used to compare simulated values with observed values. Geometric meaning of the parameters, the first and second derivative and inflection points of Gauss function was used to analysis the growing trend of ground traits of Salvia miltiorrhiza. The determined coefficient (R2) of those models were higher than0.9, and the accuracy of the prediction of these models based on TEP was higher than those models based on GOD. The results show:(1) the indexes would reach the maximum When the TEP of stem fresh weight, leaf area and leaf fresh weight reached942.1,653.6,671.9MJ/m2respectively;(2) the growth rates would be the fastest When the TEP of stem fresh weight, leaf area and leaf fresh weight reached405.3,501.2,510.2MJ/m2;(3)when the stem fresh weight, leaf area and leaf fresh weight under the TEP reached239.5,264.2,259.8MJ/m2respectively, the growth rates were slow. During this time to the TEP reach peak, the growth rates increased dramatically. Then the growth rates slowed down as time prolonging. Both the growth curves of stem fresh weight, leaf fresh weight and leaf area and Gaussian function achieve a higher level of compliance, and the models based on TEP show a better predictability than based on GOD.3. Study on the growth rhythm model of the main stem height, main stem diameter and leafage of Salvia Miltiorrhiza based on multivariate logistic equation At first, this paper ranked the impacts of impact factor (accumulated photo-radiation (PR), accumulated air humidity (AH), time and GOD) on main stem height, main stem diameter and leaf age of Salvia Miltiorrhiza respectively. Secondly, based on the most influential selected factor, the growth model had been established by the classic logistic equation. At last we used accumulated photo-radiation (PR), accumulated air humidity (AH) and GOD as independent variables, the above traits growth models had been established by multivariate logistic model. The result shows:1) for the main stem diameter, the sequence of the impact factor correlation degree is PR> time>AH> GOD; for the main stem height, the sequence of the impact factor correlation degree is PR> GOD time>AH; for the leaf age, the sequence of the impact factor correlation degree is time> GOD>PR> AH.2) The models based on classic logistic model, the determined coefficient (R2) of the data of2011were higher than0.94and those models reached significant levels (P<0.01); The models based on multivariate logistic model, the determined coefficient (R2) of the data of2011were higher than0.97and those models reached significant levels (P<0.01).3) The determined coefficient (R2) and F values of the simulated values by multivariate logistic models and observed values of2012were higher than the simulated valued by classical models, and prediction accuracy multivariate more than44.5%. Both the growth curves of main stem diameter, main stem height and leaf age and multivariate logistic model were remarkably consistent, and the models base on multivariate logistic model shows a better predictability than the one based on classic logistic model.4. Study on the models for dry matter partitioning and yield prediction of Salvia Miltiorrhiza.Using accumulated product of thermal effectiveness and PAR as independent variable, the models of allocation indexes about stem, leaf, root and ground of Salvia Miltiorrhiza and dry-matter rate about above traits had been established respectively by Fourier function. The models of flower allocation indexes had been established by quadratic exponential model. The models of flower dry-matter rate had been established by3times liner function. The determined coefficient (R2) of the data of2011were higher than0.89and those models reached significant levels (P<0.01); Depending on the models of distribution index and dry-matter rate, the fresh weight in different apparatuses are obtained, and predict the yield of Salvia Miltiorrhiza. The determined coefficient (R) of simulated values and observed values of2012were higher than0.81, and reached significant levels (P<0.01), root mean square error are relatively small.