A Process-Based Simulation Model For Cotton Growth And Development

Cotton growth and development simulation model is not only a important basement for the management decision support system but also a powerful tool for evaluation of cotton production potential, prediction of effect on climate change, intelligence cultivation and assistant breeding. A process-based simulation model for cotton growth and development was created based on reference to advanced worldwide modeling theories and framework, results field experiments, and the studies systematically analyzing and integrating cotton dynamic growth and development process and effectiveness of environmental factors with respect of present research achievements. Therefore, created model had good adaptability, also built a basement for digitalization of cotton growth and its regulation.The simulation for cotton development stages was developed with using the scale of physiological development time (PDT), based on cotton physiological and ecological process. The model selected effective function types and early maturity parameter for varieties, thoroughly considers the effectiveness factors and its interactions such as temperature, photoperiod, gene types etc. The model took PDT as cotton physiological age which achieved a better continuity as well as clearer physiological meaning than scale of development period or development index. The studies built genetic parameters with analysis of experimental data for different experiment sites, ecological zones at various years, so that all varieties could reach the same physiological time at a given development stage. With using soil temperature to simulating process of emergence, and using temperature compensation coefficient (CE) to quantify air temperature compensation effect by soil temperature increase, cotton growing with plastic film mulching could reach the same physiological time at a given development stage. Thus, the model could be adapted to plastic film mulching cultivation.The model introduced Gaussian integration method for computation of daily total canopy PAR interception and assimilation, considered not only PAR daily course but also the influence of direct PAR extinction coefficient by daily solar height course. The model calculated photosynthesis on the basis of both direct radiation and diffusion radiation from sky, canopy and soil. The effectiveness of temperature, physiological age, Nitrogen andwater status on photosynthesis and respiration were fully quantified.Cotton morphogenesis simulation was driven by daily RTE (relative therm?) effectiveness), with consideration of quantifying adverse effectiveness of temperature' which over optimal scale and temperature compensation from soil temperature increase under the plastic film coverage. The RMSE of simulated and observed value for leaf number, plant height, fruit branch and fruitnode were 0.7, 1.6 cm, 0.8 and 3.3 respectively.LAI dynamic course was simulated on the basis of the relationship among dry matter source, flow and sink. LAI was integrated on its simulating under source limitation and sink limitation. Through parameter calibration and model validation in different varieties and RMSE reached 0.3, showed a good adaptability.By integrating root sampler and image analysis methods for root sample collection in the field and root measurement, this study investigated dynamic characteristics of root growth and spatial distribution as root length, diameter and area within 0-100cm soil depth and 0-40cm width. Compared with the role measurement method, the image analysis method for root study showed advantages and credentials. The result indicated that cotton root length density (RLD) averaged 1.21-1.27 mm.cm"3 during flowering and boll stage, 1.04-1.12 mm.cm"3 after boll open, 0.76 mm.cm"3 during harvest period. Root diameter was significantly different among genotypes (P=0.022) , with the thickest root diameter of 0.52mm in insect resistant hybrid cultivar and thinnest root diameter of 0.36mm in early matured cultivar. The root diameter is not significantly different within soil layers (P=0.869), but...