Ragweed pollen source inventory for France – The second largest centre of Ambrosia in Europe

Highlights

• The first ragweed pollen inventory for France with 5 km resolution.

• Combines Ambrosia pollen data, Corine Land Cover data and a Digital Elevation Model.

• The Rhone Valley had the highest emission of ragweed pollen in France.

• Results presented as a national (French) and European infection level.

• The data are directly comparable with other such inventories in Europe.

Abstract

France, in particular the Rhône-Alpes region, is one of the three main centres of ragweed (Ambrosia) in Europe. The aim of this study is to develop a gridded ragweed pollen source inventory for all of France that can be used in assessments, eradication plans and by atmospheric models for describing concentrations of airborne ragweed pollen. The inventory combines information about spatial variations in annual Ambrosia pollen counts, knowledge of ragweed ecology, detailed land cover information and a Digital Elevation Model. The ragweed inventory consists of a local infection level on a scale of 0–100% (where 100% is the highest plant abundance per area in the studied region) and a European infection level between 0% and 100% (where 100% relates to the highest identified plant abundance in Europe using the same methodology) that has been distributed onto the EMEP grid with 5 km × 5 km resolution. The results of this analysis showed that some of the highest mean annual ragweed pollen concentrations were recorded at Roussillon in the Rhône-Valley. This is reflected by the inventory, where the European infection level has been estimated to reach 67.70% of the most infected areas in Europe i.e. Kecskemét in central Hungary. The inventory shows that the Rhône Valley is the most heavily infected part of France. Central France is also infected, but northern and western parts of France are much less infected. The inventory can be entered into atmospheric transport models, in combination with other components such as a phenological model and a model for daily pollen release, in order to simulate the dispersion of ragweed pollen within France as well as potential long-distance transport from France to other European countries.

Introduction

Predictions of atmospheric pollen concentrations are traditionally produced using statistical receptor-orientated models that are constructed without knowledge of source conditions or calculations of advection and diffusion. However, there are a number of similarities between airborne pollen concentrations and chemical air pollutants (Skjøth et al., 2008), which suggest that atmospheric transport models can be used to deal with the atmospheric dispersal of pollen in general (Skjøth et al., 2010 and references therein). Typical source-orientated models are Chemistry Transport Models (CTMs) like the EMEP model (Simpson et al., 2012), CHIMERE (de Meij et al., 2009), MATCH (Langner et al., 2009), LOTUS-EUROS (Wichink Kruit et al., 2012), and DEHM (Skjøth et al., 2011). Such source-orientated models make certain assumptions about the dispersal environment that needs to be satisfied before they can be applied and are based on known or estimated emission rates (Seinfeld and Pandis, 1998).

The performance of source-orientated models is strongly dependent on the quality of the emissions data (Sofiev et al., 2006). Emissions are among the most important input to these models (e.g. Hertel et al., 2012), and are considered to be one of the largest factors of uncertainty (e.g. Reis et al., 2009). Source based models for pollen dispersal require emission inventories of the pollen producing species, e.g. the distribution and abundance, commonly presented as the amount of area covered by certain plant species. This can be achieved by using a bottom-up approach that requires statistical data of pollen producing species (i.e. location and amount) within a given geographical area (e.g. Skjøth et al., 2008). This type of data cannot be found for many important allergenic plants, such as herbaceous weed species like Ambrosia. An alternative to the bottom-up approach is the top-down approach that uses measured pollen concentrations as a starting point and then a backwards calculation method for estimating the geographical distribution of the species of interest (e.g. Skjøth et al., 2010).

Pollen grains from the genus Ambrosia (ragweed) are very potent aeroallergens that appear to induce asthma about twice as often as other pollen (Dahl et al., 1999). Among ragweed species, only Ambrosia maritima L. is native to Europe (Hansen, 1976). Four other species have been introduced from North America. The area invaded by ragweed in a particular region is positively correlated with the length of time passed since its introduction. The most widespread and important in terms of allergy is Ambrosia artemisiifolia L. (common or short ragweed), which according to the analysis of herbarium records has been present in France since 1863 (Chauvel et al., 2006).

Each ragweed plant produces millions of pollen grains (Fumanal et al., 2007) that are small (18–22 μm) and suitable for long-distance transport (LDT) when the atmospheric conditions are favourable (Kasprzyk et al., 2011, Sikoparija et al., 2009, Smith et al., 2008, Stach et al., 2007). The most important sources of ragweed pollen in Europe are considered to be the Rhône Valley (France), parts of Northern Italy and the Pannonian Plain (Skjøth et al., 2010), where the Rhône Valley can contribute to the pollen spectra of other French Regions (Thibaudon et al., 2009, Thibaudon et al., 2010). France is also considered to be the origin of ragweed pollen recorded in Catalonia-Spain (Belmonte et al., 2000) and Switzerland (Peeters, 2000, Taramarcaz et al., 2005), thus contributing to LDT. This study aims to develop an inventory of ragweed pollen sources in France, which can be used by source-orientated models like EMEP or CHIMERE as well as in assessments and for developing eradications plans.