Variations, trends and forecast models for the airborne Olea europaea pollen season in Tétouan (NW of Morocco).

Abstract

Olea europaea L. is an emblematic tree plantation of the Mediterranean basin and one of the main sources of allergenic pollen. In this study, we examined variations in airborne Olea pollen season, trends and built forecast models based on multiple regression analysis over a 13-year period (2008-2019, 2022) in NW of Morocco (Tétouan), focusing on start date of pollination (SDP), end date of pollination (EDP), peak date (PD), and pre-peak pollen Integral (PPI). Spearman’s correlation analysis highlighted the importance of different pre-season meteorological parameters on the features of Olea pollen season depending on the period considered. SDP became earlier with increasing minimum temperature in March, while EDP was mainly influenced by precipitation in February and PD is earlier with increasing maximum temperature and precipitation in February. Linear regression results indicated a trend toward a shorter pollination period, almost significant, by delaying SDP rather than earlier EDP, probably due to the significant decrease in minimum temperature between January and April. The best regression models predicted the characteristics of the Olea pollen season to within 2 days and a value close to the PPI at 45 pollen*day/m3, and achieved an accuracy between 58 and 95%.The strongest predictors when forecasting SDP, EDP, PD and PPI were minimum temperature in March, precipitation in April, maximum temperature in February and minimum temperature in November, respectively. Findings suggest that olive reproductive cycle is considerably dependent on pre-season meteorological parameters. Further performed statistical analysis should be made to improve traditional models using a long data series.