Temporal beta diversity of lake plants is determined by concomitant changes in environmental factors across decades

Abstract

Abstract To comprehensively understand the impact of anthropogenic activities on biodiversity, we must understand how biodiversity has changed over time and what are the underlying processes. A growing body of evidence has shown that beta diversity reveals more about temporal changes in biodiversity compared with alpha diversity. Temporal beta diversity indicates, for example, degrees of change in species composition at single locations through time. We examined whether freshwater plant communities showed different patterns in temporal beta diversity in relation to concomitant changes in environmental conditions across decades. To do this, we used presence–absence data of lake plants for five decades (1940s–2010s) from southern Finland and calculated temporal beta diversity indices (TBI) for each lake between pairs of decades to the whole community, hydrophytes and helophytes. To get insights into possible processes behind the observed trends, we decomposed TBIs into beta diversity contributed by either temporal losses or temporal gains of species. We related TBIs and their loss and gain components to lake landscape position and changes in environmental variables. Based on comparisons of TBIs between the survey decade pairs, the temporal change in aquatic plant communities was modest through decades. Hydrophyte assemblages have changed more than helophyte assemblages. The main changes in temporal beta diversity occurred from the 1940s to the 1970s, when the gain of new species was the dominant process in the lakes throughout the landscape. Following that period, there was only modest changes, but from the 2000s to the 2010s, the dominant process was the loss of species. Temporal changes in environmental conditions played a key role in explaining the TBI. Our results showed that relying on only two survey points in time can result in limited knowledge of the ecological phenomenon under study and, for example, an exceptional year in terms of weather conditions can hinder detecting overall long‐term trends in compositional changes. Therefore, future studies should try to combine data from several decades to overcome the typical limitations of temporal information.