Success and failure in the invasion process of non-native agricultural weeds in Britain

Work completed during the time funded by the grant:

The first part of the project was spent to get data on non-native agricultural weeds in Britain. A dataset was compiled using the data from the New Atlas of the British and Irish Flora which was then updated with data provided by the Botanical Society of the British Isles (BSBI), an organisation of volunteer botanists systematically recording plant species’ occurrences in Britain and Ireland for over a hundred years.

This initial dataset contained information on all non-native plant species that have been recorded in agricultural habitats, the date of the first record in Britain and the number of hectads (10 km2) in which the species has been recorded from 1987-1999. From this dataset of 289 species only 15 were first recorded since 1950, the timeframe of the project.  Six of these records were from the Isles of Scilly.

To assess the impacts of these species on agricultural production the dataset of the “Encyclopaedia of arable weeds” was accessed via ADAS (Denise Ginsburg, personal communication).  This dataset contains impact ratings for 233 native and non-native agricultural weeds. In contrast to the BSBI/New Atlas dataset the ADAS list does not include species that have just a few records in Britain and perhaps just occasional occurrences in arable fields.

The first result from the combination of the two datasets was that of the species first recorded since 1950 none had been recorded in the ADAS list, indicating they were all considered not to have any impact. Native weed species were on average more widely distributed than archaeophytes (species introduced before 1500) and archaeophytes wider than neophytes (introduced since 1500). However, there were no significant differences between these three groups with regard to the impact classification.

In conclusion it seems that all recent non-native weed introductions have been “failures” up to now. One reason for that could be that a high proportion of these records were from specialised crop productions such as ornamental bulb productions in the Isles of Scilly. Indeed, some of the species are plants used as ornamentals (e.g. Narcissus, Iris, Freesia and Ixia species) that have become established in sites where they have been cultivated previously but not spread much further in similar habitats.

However, there are also species that may not yet been widespread but could be favoured by climate change. For example Echinochloa esculenta was first recorded in 1971 and is now found in 68 hectads. The species is a serious weed for example in Australia where it has significant impacts on sorghum cultivation. Similar information was gathered for the rest of the species sample.

Collaborations and future work

During the project valuable collaborations were established with the BSBI and ADAS with both showing interest in future collaboration on the topic and possible project funding applications. A more concrete plan exists to use the results in a research project proposal aiming to update and extent an existing online database of weeds currently maintained by Garden Organic. The grant was also highlighted in an enquiry to Defra about a possible horizon scanning project for potential future weeds in Britain (with Paul Neve and Jonathan Storkey) to which Defra so far has not responded.

Publications

A publication of the results is in preparation. However, because of the small number of species recorded since 1950 these results will be complemented by a comparative analysis with earlier introduced non-native and native agricultural weed species as well as with non-native plants introduced into different habitats since 1950. The aim of this comparison is to find out if arable habitats have been less susceptible to new introductions, and if time lags from the first record to a noticeable negative impact may be different in different habitats.

Katharina Dehnen-Schmutz, 1st August 2013

School of Life Sciences, University of Warwick