مدل‌سازی مراحل فنولوژیک و پارامترهای عملکردی گیاه باقلا در شرایط شرق استان گلستان (ارزیابی مدل)

در حال حاضر باقلا به‌عنوان یکی از محصولات مهم در منطقه خاورمیانه، افریقا، چین و نقاطی از اروپا و استرالیا در تغذیه انسان و دام مورد توجه قرار دارد. مدل، ابزاری است که ما را در تفسیر و درک دنیایی که در آن زندگی می‌کنیم، یاری می‌کند. پیش‌بینی دقیق فنولوژی گیاهان زراعی از ویژگی‌های ضروری مدل‌های شبیه‌سازی به‌شمار می‌رود. از مدل‌ها می‌توان در بهبود مدیریت تولید گیاهان زراعی برای پیش‌بینی تاریخ‌های احتمالی برداشت یا پیش‌بینی عملکرد نهایی، یا به‌صورت فعال‌تر، برای پیش‌بینی مطمئن زمان وقوع حوادث فنولوژیکی استفاده نمود. مدل استفاده‌شده در این مطالعه، مدل ارائه‌شده توسط سلطانی و همکاران برای نخود بود. برای پارامترهای ورودی مدل، از مقادیر به‌دست‌آمده از آزمایش‌های مزرعه‌ای در دو سال زراعی91-1390 و 92-1391 که در مزرعه تحقیقاتی دانشکده علوم کشاورزی و منابع طبیعی دانشگاه گنبد کاووس در شرایط مطلوب دیم و عاری از آفات، علف‌های هرز و بیماری‌های باقلا اجرا گردیده بود، استفاده شد. 24سناریوی مختلف شامل ترکیبات مختلفی از تراکم و تاریخ کاشت در دو سال زراعی برای مدل تعریف گردید. نتایج نشان داد که مدل قادر است مراحل فنولوژیک گیاه باقلا را به‌جُز روز تا سبزشدن به‌خوبی پیش‌بینی کند. این پیش‌بینی در مراحل منتهی به برداشت باقلا از بالاترین دقت برخوردار بود، به‌نحوی‌که بالاترین ضریب تبیین با مقدار0.91 مربوط به تعداد روز تا برداشت باقلا بود. مطالعه نشان داد که مدل فوق نمی‌تواند به‌خوبی حداکثر سطح برگ، عملکرد، عملکرد بیولوژیک و شاخص برداشت را در شرایط آب و هوایی گنبد و تحت شرایط آزمایش‌های مزرعه‌ای پیش‌بینی نماید.

Modeling of phenological stages and yield parameters of faba bean in the East of Golestan province conditions

IntroductionBean is now considered as one of the most important products in the Middle East, Africa, China and parts of Europe and Australia in human and animal nutrition (Turpin et al., 2002). Bean is a suitable herb in rotation with cereals. Because it can break the cycle of many common diseases of grains and reduce the population of nematodes. The model is a tool that helps us interpret and understand the world in which we live (Boote et al., 2003). Precise prediction of the phenology of crops is one of the essential features of simulation models. From models, improved crop production management can be used to predict the probable dates of harvesting or predicting final yields, or more actively, to predict the timing of phenological events. The production and distribution of dry matter in crop simulation models is largely regulated by the timing of development stages. Understanding how the bean apos،s phenological development response to day length and temperature, creates a model for predicting the time of occurrence of different stages of development of this crop. The aim of this study was to obtain a model for predicting phenological stages, maximum leaf area index, economic and biological yield, as well as harvest index of bean plant under Gonbad conditions. Materials MethodsFor the main data, the values obtained from field experiments were carried out at the research farm of the Faculty of Agricultural Sciences and Natural Resources of GonbadeKavou’s University under favorable rainfed conditions and free of pests, weeds and bean diseases. The experiment was a factorial based on randomized complete block design with four replications in two years. Planting density was at four levels (8, 12, 16 and 20 plants/m2) and sowing date was in three levels (13 November, 1391, 13 December, 1391, 13 December 2012 and 7 December 2011, December 4, 2011 and December 17, 2011). The model used in this study was similar to that of Soltani et al, (1999) for chickpea. Meteorological and climatic data was from 1994 to 2014 of Gonbad. 24 different scenarios including different combinations of density and planting dates were defined for two years in the model. In order to compare simulated data with field experiment data, evaluation indices, explanation factor (R2), root mean square error, normalized root mean square error and line 1: 1 were used. The higher the R2 value from the linear regression of functions and the 1: 1 line closer to one, indicates a high correlation between simulated and observed values and is a sign of a better description of the model in simulating the desired attribute. Results DiscussionThe coefficient of explanation between the predicted values and obtained from field experiments on day to flowering was 0.89. The root mean square error and its normalized value were 2.99 and 1.29 days, respectively. Also, comparison of the number of days to harvest obtained from field experiments with values obtained from the model showed that all of these values are in the range of 15% up and down the 1:1 line. The results showed that the model also well predicted day to pod setting. So that the explanation coefficient, root mean square error and normalized root mean square error, between the data obtained from field experiments and the predicted data of the model were equal to 0.81, 2.99, and 12.0, respectively. Regarding the day to the physiological maturity, the coefficient of explanation, root mean square error and normalized root mean square error were 0.9, 5.84, and 12.0, respectively, which all represent a good prediction of the model. The explanation coefficient (R2) was obtained by analyzing the linear regression of the functions between day to harvest and the predicted values equal to 0.91.ConclusionThe results showed that the model could well predict the phenological stages of the faba bean, with the exception of the day to the emergence. This prediction was at the highest level in the stages leading to bean harvest, so that the highest coefficient of explanation with a value of 0.91 was found for the number of days to bean harvest. Our study showed that the above model cannot predict well the maximum leaf area index, yield, biological yield and harvest index in dune weather conditions and under our farm experiments.