The role of temperature in the onset of the Olea europaea L. pollen season in southwestern Spain

Abstract

Temperature is one of the main factors affecting the flowering of Mediterranean trees. In the case of Olea europaea L., a low-temperature period prior to bud development is essential to interrupt dormancy. After that, and once a base temperature is reached, the plant accumulates heat until flowering starts. Different methods of obtaining the best-forecast model for the onset date of the O. europaea pollen season, using temperature as the predictive parameter, are proposed in this paper. An 18-year pollen and climatic data series (1982-1999) from Cordoba (Spain) was used to perform the study. First a multiple-regression analysis using 15-day average temperatures from the period prior to flowering time was tested. Second, three heat-summation methods were used, determining the the quantities heat units (HU): accumulated daily mean temperature after deducting a threshold, growing degree-days (GDD): proposed by Snyder [J Agric Meteorol 35:353-358 (1985)] as a measure of physiological time, and accumulated maximum temperature. In the first two, the optimum base temperature selected for heat accumulation was 12.5 degrees C. The multiple-regression equation for 1999 gives a 7-day delay from the observed date. The most accurate results were obtained with the GDD method, with a difference of only 4.7 days between predicted and observed dates. The average heat accumulation expressed as GDD was 209.9 degrees C days. The HU method also gives good results, with no significant statistical differences between predictions and observations.

Influence of meteorological parameters onOleaflowering date and airborne pollen concentration in four regions of Portugal

“…In our study, average and minimum temperatures observed from January through March were among the most important independent variables to predict the total APC and the onset date of the Olea MPS. Similar results were also observed for Cordoba (Galán et al 2001;Vázquez et al 2003) although only with the average mean temperatures of January and March which were found to be the best variables for forecasting total APC. During this period the flower buds start their development leading to inflorescence growth and anther formation.…”

“…The values of R 2 and R 2 adj obtained in our study are higher than the ones obtained for Cartagena when using the regression models developed in Spain, (Moreno-Grau et al 2000) but very similar to those obtained in Có rdoba (Galán et al 2001;Vázquez et al 2003). The non-existence of a forecasting model for the total APC in the BRA and the lower R 2 obtained for the combination of all regions could be due to the influence of meteorological parameters.…”

“…Many authors have used the accumulative temperature values to predict the onset of the MPS (Galán et al 2001;Moriondo et al 2001;Vázquez et al 2003;Rodríguez-Rajo et al 2004). In our study, the predictive models considered as the best independent variables both cumulative values and average temperature values of the months prior to flowering and cumulative precipitation values observed during the autumn-winter season.…”

“…In addition, some inter-annual variations among and within regions of the total annual pollen counts and the initial, peak and ending dates of the MPS were observed. The relationship between Olea MPS features (initial, peak and total concentration) and the meteorological conditions observed prior and during this season have been widely studied (Galán et al 2001;Moriondo et al 2001;Hidalgo et al 2002;Rodríguez-Rajo et al 2004). Nevertheless, these kinds of studies are nonexistent in Portugal and often the predictive models developed for other countries can give poor results when they are applied to other regions than the ones for which they were computed.…”

May the definition of pollen season influence aerobiological results?

“…Pollen data can be used as a phenological indicator of flowering, and many phenological studies using pollen data have been conducted in order to determine the chilling and heat requirements for flowering in various species and geographical areas (Garcia-Mozo et al, 2000;Jato et al, 2000;Fornaciari, et al, 2000a;Gala´n et al, 2001;Laaidi, 2001). Meteorological data are usually supplied by local weather stations to enable calculation of thermic phenological indexes; accurate determination of the start of the Gala´n et al (1995) and Lejoly-Gabriel and Leuschner (1983) in 2002 and according to the criteria of Gala´n et al (1995) and Spieksma et al (1995) Feher and Ja`rai-Komlo´di (1997) and Lejoly-Gabriel and Leuschner (1983) in 2002 and according to the criteria of Feher and Ja`rai-Komlo´di (1997) and Spieksma et al (1995Spieksma et al ( ) in 2003 pollen season according to local flowering is therefore essential.…”