Meta - Models Efficiency in Assessing the Vegetal Pathogens Attack

The use of meta-models approach as component of the prevention strategy against different pathogens attack, specific for vegetal cultures, receives an increasing interest. Due to the importance of  the  climatic  influence  upon  pathogens  action,  almost  all  these  approaches  use  the  inputs  from meteorological stations placed in field. Another important component of such a system is represented by  a  specific  software  program  used  for  estimation  of  virtual  geographical  distribution  of  target pathogen/pathogens.  This  article  aims  to  emphasize  the possibility  of  using  the  meta-models  for predicting Phytophtora infestans Mont.  de  Bary  attack. The research was  carried  on  during  spring - autumn 2012 in a potato field located in Jucu village, on the experimental field of the UASMV Cluj - Napoca.  Climatic  conditions  (temperature  and  rainfall  regimen)  and  attack  degree  were  monitored using  a  meteorological  station  placed  on  the  field,  and  observations.  Rainfall  regimen  and  average daily  temperature  were taken  into  consideration,  in  approaching  the  multiregression  model  with  tow inputs. The  statistical  data  processing  was  performed  with  STATISTICA  7.0  v.  programme.  The analysis  of  variance  (ANOVA)  applied  to  the  interaction  between  attack  degree  of Phytophthora infestans  Mont.  de  Bary  in  potato -  temperature -  rainfall  allows  us  to  use  the  proposed multiregression  model  (F  =  7,892,  p  <  0.001).  The  multiregression  analyze  emphasize  a  strong multiple correlation coefficient of 0.895 which accounts from 81.10% of the predicted attack degree. The  regression  line,  Y  =    24.311  +  2.813X1  +  0.122X2  show,  in  climatic  conditions  specific  for  the experimental  year  2012,    the  bigger  influence  (more  than  twice)  of  the  thermic  conditions  on  the Phytophthora infestans Mont. de Bary attack degree in potato, compared to rainfall influence on the same pathogen.Obtaining the regression parameters allows us to build a meta-model for predicting the evolution of the Phytophthora infestans Mont. de Bary attack degree in potato, reliable in a high share, 81.10%, respectively, and presenting the advantage that it gives comparable results, which contribute to increase the consistency of assessments patterns.