Assessing the value of field margins for butterflies and plants: how to document and enhance biodiversity at the farm scale

With ongoing biodiversity loss in agricultural landscapes, there is an increasing demand to document how farmers preserve and enhance biodiversity on their farmland. This subject is not only of interest for conservation authorities and NGOs, food companies also look for ways to integrate biodiversity issues into their corporate activities. They want to know how their farm product suppliers contribute to biodiversity on their land. However, species counting and mapping on contracted farms seems unrealistic to these companies. Therefore, we aimed to devise simple and easy-to-use but scientifically sound parameters to assess the biodiversity value of farmlands. For this we focused on estimating the value of field margins for butterflies and typical vascular grassland plants. We identified 13 parameters that are of likely importance for the species numbers of butterflies and 14 parameters likely to be important for the species numbers of plants on field margins. These parameters were tested on a total of 70 field margins on seven farms located throughout Germany. Automatic linear modelling procedures selected six parameters as the most important variables in predicting butterfly numbers: the landscape heterogeneity of the surroundings, the time of mowing, the width and length of the margin, the grass-herb-ratio and the management of the adjacent field. For predicting plant species numbers, the automatic linear modelling procedures again selected six parameters as best predictors: the length of the margin, the presence of trees and/or shrubs, the amount of source habitats in the surroundings, the width of the margin, the nutrient availability and the landscape heterogeneity of the surroundings. The adapted final model for butterfly species numbers explained 63% of the variation. The adapted final model for typical vascular grassland plant species numbers, which excluded the margin width variable as it did not prove stable in cross-validation procedures, explained 67% of the variation in plant species numbers. Both models can be used by farmers and food companies alike to rapidly assess the present value of field margins for butterflies and plants on their (contracted) farms and to identity potential and appropriate measures to enhance biodiversity.