A novel approach to disentangle climate change and arable land management change as drivers of weed flora shifts - Combining species distribution modelling and dynamic population modelling

点击翻译按钮获取中文摘要

Arable weed distributions are influenced by environmental factors and crop management. Climate change and intensified arable land management have been important drivers of recent shifts in weed flora. Indirect impacts are also expected from on-farm adaptation measures to climate change. Although both drivers have been intensively studied, their interaction and combined effects have until now hardly been addressed. One factor may increase or decrease the impact of the other and the interactions may have spatial or temporal patterns. A major obstacle for analysis is that both act on drivers differing temporal and spatial scales, so their impacts need to be analysed with different techniques. Species distribution modelling (SDM) studies have for example predicted range shifts under future climate with a North Eastern direction. SDM can be adapted to some extent to the arable context with its regular, strong disturbances by crop measures. However, because SDM have relatively coarse resolutions they cannot incorporate local scale effects of changes in crop management such as a more simplified crop rotation.This project therefore proposes a novel, combined approach of SDM and dynamic crop:weed modelling to analyse the relative contribution of climate change and land management change on arable weed flora changes. It is hypothesised that diverse cropping systems (with diverse rotations and diverse crop management) would facilitate weed management under the impact of climate change by reducing shifts to a more homogenous weed flora composition with high abundance of single species. The project will focus on Mecklenburg-Vorpommern, a typical temperate arable region of Central Europe, but the approach will be transferable to other regions. Changes will be investigated for the time periods 2030-2060 and 2070-2100.The project addresses a number of research questions and hypotheses in three parts. First, SDM are built for weed species that are important or are expected to become important to estimate future regional distributions and species pools. The second part FlorSys is used, a very detailed crop:weed model for population dynamics of multi-species weed associations. The transferability of the model to the conditions of Northern Germany will be tested. Then the development of the species pools resulting from SDM is modelled in different scenarios of land management and climate change. Subsequently, the results of both modelling approaches are combined in ordination, clustering and regression analyses to investigate the impact of each driver and their interactions on weed floras.Besides the insight into the relation between climate change, land management change and weed floras, the approach will provide farmers with information on possible future weed problems and their causes on an appropriate, regional scale. It therefore contributes to the preparation of adaptation strategies for climate change.