Linking patterns to processes: Using hunting bag data to classify raccoon (Procyon lotor) invasion stages in Germany since the 2000s

• The invasion process of Procyon lotor in Germany started in the 1930s. • We categorise time series of raccoon hunting data over 21 years at district level. • These categories are assigned to phases of the invasion process. • We found absence, introduction, slow growth, expansion, saturation, stabilisation. • Spatial autocorrelation and two hotspots that show saturation were identified. Effective and successful management of invasive alien species requires a deep understanding of population dynamics in the invasion process. Following general concepts in invasion ecology we adapted the theoretical background to the stages of raccoon invasion in Germany. Hunting bag data provides time series over 21 years for each of 398 districts in Germany. In some cases the invasion processes might develop for longer time period than what could be studied. However it was possible to interpret individual time series as a part of the overall process: in particular absence (stage I), a lag phase – low level and strongly driven by chance (stage II), the spread phase subdivided according to the shape (linear – IIIa; exponential – IIIb and saturated – IIIc) and the stabilisation phase (stage IV) with random fluctuation at a high level. We applied a piecewise regression algorithm to differentiate between stages in the spread phase. The spatial patterns of this classification indicated two hotspots of late stages that coincide with the two assumed founding areas where raccoons ( Procyon lotor ) were first introduced, intentionally and accidentally, in the 1930s and 1940s. These results support the assumption that invasion proceeds in these phases. Densely populated areas in late phases may act as sources of immigration. Individuals may migrate to neighboring areas where they initiate or support the invasion process. Of course, this does not mean that all immigrating raccoons originate from hotspot regions in late stages of invasion. Germany hosts one of the largest non-native raccoon populations and thus represents a potential source of immigration to neighbouring countries. This study develops a spatial classification framework that can serve as an indicator for invasion dynamics in other species/ecosystems. Similar to the immigration from the two hotspots in central and north-eastern Germany into neighbouring areas, the raccoon can spread from Germany, the largest non-native population, into neighbouring countries where successive stages of invasion can be established.