Abstract:
On one hand, the main research of the presented study is the area of Palynology; on the other hand it is the field of Phenology. Analyzed within the scope of the defined investigations is pollen flight data collected by the “Stiftung Deutscher Polleninformationsdienst“ (PID) for the purpose of allergological forecasts. This data concerns the use of one or more taxa as an instrument of crop yield prediction and prediction of crop yield parameters of useful plants at an early stage in the year. Additionally, the identification and recognition of possible changes in Phenology and vegetation is of further interest. Particularly convenient is the fact that pollenflors and vegetation communities react directly to external influences. Twenty-five taxa of trees and herbal plants in total have been considered for this research. The project research area consists of selected regions and localities with Baden-Wurttemberg/SW-Germany. Pollen flight data from pollen traps at 7 different localities have been statistically evaluated and examined for correlations and temporal trends over a time span of up to 17 consecutive years. In addition to the pollen flight data, climate- and crop yield parameter data obtained by the “Deutscher Wetterdienst” (DWD) and the “Statistisches Landesamt Baden-Württemberg” (STALA BW) over several years have been integrated into the study. Furthermore, at one single location of the “Landwirtschaftlichen Versuchsanstalt für Wein- und Obstbau” (LVWO), phenological data of diverse wine varieties were examined for possible changes and trends concerning their phase entry dates. In the final research step, the pollen input at the individual pollen stations was evaluated with special regard to the surrounding vegetation and/or to the land use in order to reveal possible correlations or patterns. The identification, visualization and representation of the coherences take place using selected methods. The statistical analyzes conducted by SPSS© 12.0.1 (in regards to pollen data, crop yield data, climate data and phenological data as) consists of simple as well as multiple linear regressions. Furthermore, the analysis of the phenological data is carried out with EXCEL and with ArcGIS 3.1 & 8/ArcMAP for the land use data of the CORINE land use map. The results suggest that pollen flight data derived from pollen traps are suited for predicting crop yield parameters. Yet, this is valid on a regional scale and in depends on the individual wine varieties being observed. Both correlations of pollen and climate data and wine and climate data result in fewer, yet, at the same time, significant coherences compared to the correlations between pollen and wine data. Thus, pollen data proves to be an integrating factor of the diverse actuating variables like e.g. of climate and soil conditions affecting the crop yield parameters. Therefore, insights gained from this project promote new application possibilities for Palynology, especially in the field of crop yield prediction. Depending on the analyzed taxon, the phenological investigations of the selected pollenflors result in an up- and/or downward trend of the pollensum and the days with pollen flight. There can be seen in the individually examined wine varieties a significant, two to three week premature phenological phase arrival. This trend has also become evident in vegetation and/or yearly cylces as a result of the climate change observed over the past decades.