Study On Rapeseed Growth Simulation And Decision Support System

The research and application of crop growth model and decision support system for rapeseed management would be important for facilitating development of informational and igital agriculture. In the present study, the relationships of growth and development to environment factors were analyzed and integrated by using the field experiments data with different genotypes, sowing dates and nitrogen application levels. By adopting studying advanced modeling technology abroad and the methodology of crop growth model developed by our lab, a physiological process-based rapeseed simulation model was developed through the system analysis and mathematical modeling. By using object-oriented framework structure and component design, a component based rapeseed growth model (RapeGrow) was developed with COM standard. Then, a growth model-based decision support system for rapeseed management (GMDSSRSM) was established by integrating decision support techniques, which realized the functions of dynamic simulation, growth prediction, decision-making in rapeseed production under the environments of stand-alone and network, respectively. The present study should be useful for prediction of growth performance under different conditions and construction of rapeseed digital management system in rapeseed crop.Based on the eco-physiological processes of rapeseed development, a simulation model for predicting phenological stages of oilseed rape (brassica napus L.) was developed with the physiological development time (PDT) as time scale and by quantifying the effects of temperature (including vernalization) and photoperiod on oilseed rape. The interaction among daily thermal effectiveness, photoperiod effectiveness, vernalization effectiveness and filling rate determined the daily physiological effectiveness, which was accumulated to obtain physiological development time. Four specific genetic parameters as temperature sensitivity, physiological vernalization time, photoperiod sensitivity and basic filling duration factor were added to adjust the genotypic differences in rapeseed development so that all different varieties could reach the same physiological development time at a given development stage.Dynamic leaf area expansion in rapeseed was simulated through the relationship between LAI and source or sink limited dry matter. The daily leaf expansion rate increased exponentially under the source limitation, as influenced by the factors of temperature, water and nitrogen deficits, whereas under the sink limitation, leaf area expansion was quantified on the basis of specific leaf area (SLA). Pod area was calculated from specific pod area (SPA) and pod dry matter. The specific leaf area and specific pod area were determined from physiological development time (PDT) and genotype.Based on the characters of canopy structure, a simulation model for photosynthetic production and dry matter accumulation was established for rapeseed crop. The model used a "3 layer model", calculating the radiation interception and photosynthesis on the layers of flowers, pods and leaves, respectively. Gaussian integration method was used to calculate the photosynthesis of each layer, and the daily total canopy photosynthesis was the sum of the each layer's photosynthesis. The effects of physiological age, temperature, nitrogen and water deficit factors on maximum photosynthesis were quantified respectively. The model also considered maintenance and growth respiration in determining net photosynthetic production.Based on eco-physiological processes, a simulation model on dry matter partitioning and yield formation in rapeseed was developed through the relationships of partitioning index to development progress and environment factors. Partitioning indices of green leaf, stem and pod changed with physiological development time, as influenced by genotype, sowing date, nitrogen level and water status. Green leaf dry weight was regulated by nitrogen nutrition index (NNI), the higher nitrogen nutrition index and the higher partitioning index of green leaf. The factor of sowing date was introduced to regulate partitioning index of pod, which increased with sowing date delayed.By integrating the above submodels with existing wate and nutrient sub-models, a comprehensimve growth simulation model (RapeGrow) was constructed for rapeseed crop. The RapeGrow includes six sub models for simulating phasic development, organ formation, biomass production, yield and quality formation, soil-crop water relations, and nutrient (N, P, K) balance. Validation of these individual sub-models with the field experiments data of different genotypes, sowing dates, and nitrogen levels showed that the models could accurately predict main development stages, green area index, dry matter accumulation and partitioning to different organs under different conditions. The present model appears to have good explanation, reliable prediction and wide adaptation.Based on the analysis and extraction of research data on cultural theories and technologies in winter rapeseed, a quantitive and general knowledge model was developed for present design of suitable green area index dynamics in winter rapeseed. The model was driven by physiological development time-based growing degree days and by quantitative relationships of growth characters to variety traits, eco-environments and production levels, and can be used for designing the suitable time-course dynamics of leaf area index and pod area index under different conditions. Case studies with the data sets of normal climatic year, different variety types and yield levels at different eco-sites indicate a good performance of the model system in guidance and wide applicability.Driven by soil, variety, weather and management databases, a growth model-based decision system for rapeseed management (GMDSSRSM) was developed by integrating growth simulation technology and component-based software technology, including the stand-alone edition (GMDSSRSM~A) and web-based edition (GMDSSRSM~W). The GMDSSRSM~A was established on the platforms of VC++ and VB by adopting the characteristics of object-oriented and component-based software and coupling the growth prediction function and decision-making function for cultural management. Then, GMDSSRSM~W was further developed on the platform of. net with the language of C# for web users. The implemented system can be used for evaluating individual and comprehensive management strategies based on the results of rapeseed growth simulation under various environments and different genotypes. The GMDSSRSM should quantitivelay a technical foundation for construction and application of digital farming system in modern rapeseed production.